Proceedings of the 12th Language Resources and Evaluation Conference

Nicoletta Calzolari, Frédéric Béchet, Philippe Blache, Khalid Choukri, Christopher Cieri, Thierry Declerck, Sara Goggi, Hitoshi Isahara, Bente Maegaard, Joseph Mariani, Hélène Mazo, Asuncion Moreno, Jan Odijk, Stelios Piperidis (Editors)

Anthology ID:
Marseille, France
European Language Resources Association
Bib Export formats:

pdf bib
Proceedings of the 12th Language Resources and Evaluation Conference
Nicoletta Calzolari | Frédéric Béchet | Philippe Blache | Khalid Choukri | Christopher Cieri | Thierry Declerck | Sara Goggi | Hitoshi Isahara | Bente Maegaard | Joseph Mariani | Hélène Mazo | Asuncion Moreno | Jan Odijk | Stelios Piperidis

pdf bib
Neural Mention Detection
Juntao Yu | Bernd Bohnet | Massimo Poesio

Mention detection is an important preprocessing step for annotation and interpretation in applications such as NER and coreference resolution, but few stand-alone neural models have been proposed able to handle the full range of mentions. In this work, we propose and compare three neural network-based approaches to mention detection. The first approach is based on the mention detection part of a state of the art coreference resolution system ; the second uses ELMO embeddings together with a bidirectional LSTM and a biaffine classifier ; the third approach uses the recently introduced BERT model. Our best model (using a biaffine classifier) achieves gains of up to 1.8 percentage points on mention recall when compared with a strong baseline in a HIGH RECALL coreference annotation setting. The same model achieves improvements of up to 5.3 and 6.2 p.p. when compared with the best-reported mention detection F1 on the CONLL and CRAC coreference data sets respectively in a HIGH F1 annotation setting. We then evaluate our models for coreference resolution by using mentions predicted by our best model in start-of-the-art coreference systems. The enhanced model achieved absolute improvements of up to 1.7 and 0.7 p.p. when compared with our strong baseline systems (pipeline system and end-to-end system) respectively. For nested NER, the evaluation of our model on the GENIA corpora shows that our model matches or outperforms state-of-the-art models despite not being specifically designed for this task.

pdf bib
Model-based Annotation of Coreference
Rahul Aralikatte | Anders Søgaard

Humans do not make inferences over texts, but over models of what texts are about. When annotators are asked to annotate coreferent spans of text, it is therefore a somewhat unnatural task. This paper presents an alternative in which we preprocess documents, linking entities to a knowledge base, and turn the coreference annotation task in our case limited to pronouns into an annotation task where annotators are asked to assign pronouns to entities. Model-based annotation is shown to lead to faster annotation and higher inter-annotator agreement, and we argue that it also opens up an alternative approach to coreference resolution. We present two new coreference benchmark datasets, for English Wikipedia and English teacher-student dialogues, and evaluate state-of-the-art coreference resolvers on them.

pdf bib
Cross-lingual Zero Pronoun Resolution
Abdulrahman Aloraini | Massimo Poesio

In languages like Arabic, Chinese, Italian, Japanese, Korean, Portuguese, Spanish, and many others, predicate arguments in certain syntactic positions are not realized instead of being realized as overt pronouns, and are thus called zero- or null-pronouns. Identifying and resolving such omitted arguments is crucial to machine translation, information extraction and other NLP tasks, but depends heavily on semantic coherence and lexical relationships. We propose a BERT-based cross-lingual model for zero pronoun resolution, and evaluate it on the Arabic and Chinese portions of OntoNotes 5.0. As far as we know, ours is the first neural model of zero-pronoun resolution for Arabic ; and our model also outperforms the state-of-the-art for Chinese. In the paper we also evaluate BERT feature extraction and fine-tune models on the task, and compare them with our model. We also report on an investigation of BERT layers indicating which layer encodes the most suitable representation for the task.

pdf bib
Affection Driven Neural Networks for Sentiment Analysis
Rong Xiang | Yunfei Long | Mingyu Wan | Jinghang Gu | Qin Lu | Chu-Ren Huang

Deep neural network models have played a critical role in sentiment analysis with promising results in the recent decade. One of the essential challenges, however, is how external sentiment knowledge can be effectively utilized. In this work, we propose a novel affection-driven approach to incorporating affective knowledge into neural network models. The affective knowledge is obtained in the form of a lexicon under the Affect Control Theory (ACT), which is represented by vectors of three-dimensional attributes in Evaluation, Potency, and Activity (EPA). The EPA vectors are mapped to an affective influence value and then integrated into Long Short-term Memory (LSTM) models to highlight affective terms. Experimental results show a consistent improvement of our approach over conventional LSTM models by 1.0 % to 1.5 % in accuracy on three large benchmark datasets. Evaluations across a variety of algorithms have also proven the effectiveness of leveraging affective terms for deep model enhancement.

pdf bib
Linguistic, Kinematic and Gaze Information in Task Descriptions : The LKG-CorpusLKG-Corpus
Tim Reinboth | Stephanie Gross | Laura Bishop | Brigitte Krenn

Data from neuroscience and psychology suggest that sensorimotor cognition may be of central importance to language. Specifically, the linguistic structure of utterances referring to concrete actions may reflect the structure of the sensorimotor processing underlying the same action. To investigate this, we present the Linguistic, Kinematic and Gaze information in task descriptions Corpus (LKG-Corpus), comprising multimodal data on 13 humans, conducting take, put, and push actions, and describing these actions with 350 utterances. Recorded are audio, video, motion and eye-tracking data while participants perform an action and describe what they do. The dataset is annotated with orthographic transcriptions of utterances and information on : (a) gaze behaviours, (b) when a participant touched an object, (c) when an object was moved, (d) when a participant looked at the location s / he would next move the object to, (e) when the participant’s gaze was stable on an area. With the exception of the annotation of stable gaze, all annotations were performed manually. With the LKG-Corpus, we present a dataset that integrates linguistic, kinematic and gaze data with an explicit focus on relations between action and language. On this basis, we outline applications of the dataset to both basic and applied research.

pdf bib
The ACQDIV Corpus Database and Aggregation PipelineACQDIV Corpus Database and Aggregation Pipeline
Anna Jancso | Steven Moran | Sabine Stoll

We present the ACQDIV corpus database and aggregation pipeline, a tool developed as part of the European Research Council (ERC) funded project ACQDIV, which aims to identify the universal cognitive processes that allow children to acquire any language. The corpus database represents 15 corpora from 14 typologically maximally diverse languages. Here we give an overview of the project, database, and our extensible software package for adding more corpora to the current language sample. Lastly, we discuss how we use the corpus database to mine for universal patterns in child language acquisition corpora and we describe avenues for future research.

pdf bib
Understanding the Dynamics of Second Language Writing through Keystroke Logging and Complexity Contours
Elma Kerz | Fabio Pruneri | Daniel Wiechmann | Yu Qiao | Marcus Ströbel

The purpose of this paper is twofold : [ 1 ] to introduce, to our knowledge, the largest available resource of keystroke logging (KSL) data generated by Etherpad (, an open-source, web-based collaborative real-time editor, that captures the dynamics of second language (L2) production and [ 2 ] to relate the behavioral data from KSL to indices of syntactic and lexical complexity of the texts produced obtained from a tool that implements a sliding window approach capturing the progression of complexity within a text. We present the procedures and measures developed to analyze a sample of 14,913,009 keystrokes in 3,454 texts produced by 512 university students (upper-intermediate to advanced L2 learners of English) (95,354 sentences and 18,32,027 words) aiming to achieve a better alignment between keystroke-logging measures and underlying cognitive processes, on the one hand, and L2 writing performance measures, on the other hand. The resource introduced in this paper is a reflection of increasing recognition of the urgent need to obtain ecologically valid data that have the potential to transform our current understanding of mechanisms underlying the development of literacy (reading and writing) skills.

pdf bib
Developing a Corpus of Indirect Speech Act Schemas
Antonio Roque | Alexander Tsuetaki | Vasanth Sarathy | Matthias Scheutz

Resolving Indirect Speech Acts (ISAs), in which the intended meaning of an utterance is not identical to its literal meaning, is essential to enabling the participation of intelligent systems in peoples’ everyday lives. Especially challenging are those cases in which the interpretation of such ISAs depends on context. To test a system’s ability to perform ISA resolution we need a corpus, but developing such a corpus is difficult, especialy given the contex-dependent requirement. This paper addresses the difficult problems of constructing a corpus of ISAs, taking inspiration from relevant work in using corpora for reasoning tasks. We present a formal representation of ISA Schemas required for such testing, including a measure of the difficulty of a particular schema. We develop an approach to authoring these schemas using corpus analysis and crowdsourcing, to maximize realism and minimize the amount of expert authoring needed. Finally, we describe several characteristics of collected data, and potential future work.

pdf bib
MAGPIE : A Large Corpus of Potentially Idiomatic ExpressionsMAGPIE: A Large Corpus of Potentially Idiomatic Expressions
Hessel Haagsma | Johan Bos | Malvina Nissim

Given the limited size of existing idiom corpora, we aim to enable progress in automatic idiom processing and linguistic analysis by creating the largest-to-date corpus of idioms for English. Using a fixed idiom list, automatic pre-extraction, and a strictly controlled crowdsourced annotation procedure, we show that it is feasible to build a high-quality corpus comprising more than 50 K instances, an order of a magnitude larger than previous resources. Crucial ingredients of crowdsourcing were the selection of crowdworkers, clear and comprehensive instructions, and an interface that breaks down the task in small, manageable steps. Analysis of the resulting corpus revealed strong effects of genre on idiom distribution, providing new evidence for existing theories on what influences idiom usage. The corpus also contains rich metadata, and is made publicly available.

pdf bib
Effort Estimation in Named Entity Tagging Tasks
Inês Gomes | Rui Correia | Jorge Ribeiro | João Freitas

Named Entity Recognition (NER) is an essential component of many Natural Language Processing pipelines. However, building these language dependent models requires large amounts of annotated data. Crowdsourcing emerged as a scalable solution to collect and enrich data in a more time-efficient manner. To manage these annotations at scale, it is important to predict completion timelines and compute fair pricing for workers in advance. To achieve these goals, we need to know how much effort will be taken to complete each task. In this paper, we investigate which variables influence the time spent on a named entity annotation task by a human. Our results are two-fold : first, the understanding of the effort-impacting factors which we divided into cognitive load and input length ; and second, the performance of the prediction itself. On the latter, through model adaptation and feature engineering, we attained a Root Mean Squared Error (RMSE) of 25.68 words per minute with a Nearest Neighbors model.

pdf bib
Constructing Multimodal Language Learner Texts Using LARA : Experiences with Nine LanguagesLARA: Experiences with Nine Languages
Elham Akhlaghi | Branislav Bédi | Fatih Bektaş | Harald Berthelsen | Matthias Butterweck | Cathy Chua | Catia Cucchiarin | Gülşen Eryiğit | Johanna Gerlach | Hanieh Habibi | Neasa Ní Chiaráin | Manny Rayner | Steinþór Steingrímsson | Helmer Strik

LARA (Learning and Reading Assistant) is an open source platform whose purpose is to support easy conversion of plain texts into multimodal online versions suitable for use by language learners. This involves semi-automatically tagging the text, adding other annotations and recording audio. The platform is suitable for creating texts in multiple languages via crowdsourcing techniques that can be used for teaching a language via reading and listening. We present results of initial experiments by various collaborators where we measure the time required to produce substantial LARA resources, up to the length of short novels, in Dutch, English, Farsi, French, German, Icelandic, Irish, Swedish and Turkish. The first results are encouraging. Although there are some startup problems, the conversion task seems manageable for the languages tested so far. The resulting enriched texts are posted online and are freely available in both source and compiled form.

pdf bib
A Dataset for Investigating the Impact of Feedback on Student Revision Outcome
Ildiko Pilan | John Lee | Chak Yan Yeung | Jonathan Webster

We present an annotation scheme and a dataset of teacher feedback provided for texts written by non-native speakers of English. The dataset consists of student-written sentences in their original and revised versions with teacher feedback provided for the errors. Feedback appears both in the form of open-ended comments and error category tags. We focus on a specific error type, namely linking adverbial (e.g. however, moreover) errors. The dataset has been annotated for two aspects : (i) revision outcome establishing whether the re-written student sentence was correct and (ii) directness, indicating whether teachers provided explicitly the correction in their feedback. This dataset allows for studies around the characteristics of teacher feedback and how these influence students’ revision outcome. We describe the data preparation process and we present initial statistical investigations regarding the effect of different feedback characteristics on revision outcome. These show that open-ended comments and mitigating expressions appear in a higher proportion of successful revisions than unsuccessful ones, while directness and metalinguistic terms have no effect. Given that the use of this type of data is relatively unexplored in natural language processing (NLP) applications, we also report some observations and challenges when working with feedback data.

pdf bib
Using Multilingual Resources to Evaluate CEFRLex for Learner ApplicationsCEFRLex for Learner Applications
Johannes Graën | David Alfter | Gerold Schneider

The Common European Framework of Reference for Languages (CEFR) defines six levels of learner proficiency, and links them to particular communicative abilities. The CEFRLex project aims at compiling lexical resources that link single words and multi-word expressions to particular CEFR levels. The resources are thought to reflect second language learner needs as they are compiled from CEFR-graded textbooks and other learner-directed texts. In this work, we investigate the applicability of CEFRLex resources for building language learning applications. Our main concerns were that vocabulary in language learning materials might be sparse, i.e. that not all vocabulary items that belong to a particular level would also occur in materials for that level, and, on the other hand, that vocabulary items might be used on lower-level materials if required by the topic (e.g. with a simpler paraphrasing or translation). Our results indicate that the English CEFRLex resource is in accordance with external resources that we jointly employ as gold standard. Together with other values obtained from monolingual and parallel corpora, we can indicate which entries need to be adjusted to obtain values that are even more in line with this gold standard. We expect that this finding also holds for the other languages

pdf bib
Toward a Paradigm Shift in Collection of Learner Corpora
Anisia Katinskaia | Sardana Ivanova | Roman Yangarber

We present the first version of the longitudinal Revita Learner Corpus (ReLCo), for Russian. In contrast to traditional learner corpora, ReLCo is collected and annotated fully automatically, while students perform exercises using the Revita language-learning platform. The corpus currently contains 8 422 sentences exhibiting several types of errorsgrammatical, lexical, orthographic, etc.which were committed by learners during practice and were automatically annotated by Revita. The corpus provides valuable information about patterns of learner errors and can be used as a language resource for a number of research tasks, while its creation is much cheaper and faster than for traditional learner corpora. A crucial advantage of ReLCo that it grows continually while learners practice with Revita, which opens the possibility of creating an unlimited learner resource with longitudinal data collected over time. We make the pilot version of the Russian ReLCo publicly available.

pdf bib
MultiWOZ 2.1 : A Consolidated Multi-Domain Dialogue Dataset with State Corrections and State Tracking BaselinesMultiWOZ 2.1: A Consolidated Multi-Domain Dialogue Dataset with State Corrections and State Tracking Baselines
Mihail Eric | Rahul Goel | Shachi Paul | Abhishek Sethi | Sanchit Agarwal | Shuyang Gao | Adarsh Kumar | Anuj Goyal | Peter Ku | Dilek Hakkani-Tur

MultiWOZ 2.0 (Budzianowski et al., 2018) is a recently released multi-domain dialogue dataset spanning 7 distinct domains and containing over 10,000 dialogues. Though immensely useful and one of the largest resources of its kind to-date, MultiWOZ 2.0 has a few shortcomings. Firstly, there are substantial noise in the dialogue state annotations and dialogue utterances which negatively impact the performance of state-tracking models. Secondly, follow-up work (Lee et al., 2019) has augmented the original dataset with user dialogue acts. This leads to multiple co-existent versions of the same dataset with minor modifications. In this work we tackle the aforementioned issues by introducing MultiWOZ 2.1. To fix the noisy state annotations, we use crowdsourced workers to re-annotate state and utterances based on the original utterances in the dataset. This correction process results in changes to over 32 % of state annotations across 40 % of the dialogue turns. In addition, we fix 146 dialogue utterances by canonicalizing slot values in the utterances to the values in the dataset ontology. To address the second problem, we combined the contributions of the follow-up works into MultiWOZ 2.1. Hence, our dataset also includes user dialogue acts as well as multiple slot descriptions per dialogue state slot. We then benchmark a number of state-of-the-art dialogue state tracking models on the MultiWOZ 2.1 dataset and show the joint state tracking performance on the corrected state annotations.

pdf bib
A Comparison of Explicit and Implicit Proactive Dialogue Strategies for Conversational Recommendation
Matthias Kraus | Fabian Fischbach | Pascal Jansen | Wolfgang Minker

Recommendation systems aim at facilitating information retrieval for users by taking into account their preferences. Based on previous user behaviour, such a system suggests items or provides information that a user might like or find useful. Nonetheless, how to provide suggestions is still an open question. Depending on the way a recommendation is communicated influences the user’s perception of the system. This paper presents an empirical study on the effects of proactive dialogue strategies on user acceptance. Therefore, an explicit strategy based on user preferences provided directly by the user, and an implicit proactive strategy, using autonomously gathered information, are compared. The results show that proactive dialogue systems significantly affect the perception of human-computer interaction. Although no significant differences are found between implicit and explicit strategies, proactivity significantly influences the user experience compared to reactive system behaviour. The study contributes new insights to the human-agent interaction and the voice user interface design. Furthermore, we discover interesting tendencies that motivate futurework.

pdf bib
Construction and Analysis of a Multimodal Chat-talk Corpus for Dialog Systems Considering Interpersonal Closeness
Yoshihiro Yamazaki | Yuya Chiba | Takashi Nose | Akinori Ito

There are high expectations for multimodal dialog systems that can make natural small talk with facial expressions, gestures, and gaze actions as next-generation dialog-based systems. Two important roles of the chat-talk system are keeping the user engaged and establishing rapport. Many studies have conducted user evaluations of such systems, some of which reported that considering the relationship with the user is an effective way to improve the subjective evaluation. To facilitate research of such dialog systems, we are currently constructing a large-scale multimodal dialog corpus focusing on the relationship between speakers. In this paper, we describe the data collection and annotation process, and analysis of the corpus collected in the early stage of the project. This corpus contains 19,303 utterances (10 hours) from 19 pairs of participants. A dialog act tag is annotated to each utterance by two annotators. We compare the frequency and the transition probability of the tags between different closeness levels to help construct a dialog system for establishing a relationship with the user.

pdf bib
The JDDC Corpus : A Large-Scale Multi-Turn Chinese Dialogue Dataset for E-commerce Customer ServiceJDDC Corpus: A Large-Scale Multi-Turn Chinese Dialogue Dataset for E-commerce Customer Service
Meng Chen | Ruixue Liu | Lei Shen | Shaozu Yuan | Jingyan Zhou | Youzheng Wu | Xiaodong He | Bowen Zhou

Human conversations are complicated and building a human-like dialogue agent is an extremely challenging task. With the rapid development of deep learning techniques, data-driven models become more and more prevalent which need a huge amount of real conversation data. In this paper, we construct a large-scale real scenario Chinese E-commerce conversation corpus, JDDC, with more than 1 million multi-turn dialogues, 20 million utterances, and 150 million words. The dataset reflects several characteristics of human-human conversations, e.g., goal-driven, and long-term dependency among the context. It also covers various dialogue types including task-oriented, chitchat and question-answering. Extra intent information and three well-annotated challenge sets are also provided. Then, we evaluate several retrieval-based and generative models to provide basic benchmark performance on the JDDC corpus. And we hope JDDC can serve as an effective testbed and benefit the development of fundamental research in dialogue task.

pdf bib
Cheese ! : a Corpus of Face-to-face French Interactions. A Case Study for Analyzing Smiling and Conversational HumorFrench Interactions. A Case Study for Analyzing Smiling and Conversational Humor
Béatrice Priego-Valverde | Brigitte Bigi | Mary Amoyal

Cheese ! is a conversational corpus. It consists of 11 French face-to-face conversations lasting around 15 minutes each. Cheese ! is a duplication of an American corpus (ref) in order to conduct a cross-cultural comparison of participants’ smiling behavior in humorous and non-humorous sequences in American English and French conversations. In this article, the methodology used to collect and enrich the corpus is presented : experimental protocol, technical choices, transcription, semi-automatic annotations, manual annotations of smiling and humor. An exploratory study investigating the links between smile and humor is then proposed. Based on the analysis of two interactions, two questions are asked : (1) Does smile frame humor? (2) Does smile has an impact on its success or failure? If the experimental design of Cheese ! has been elaborated to study specifically smiles and humor in conversations, the high quality of the dataset obtained, and the methodology used are also replicable and can be applied to analyze many other conversational activities and other multimodal modalities.

pdf bib
The Margarita Dialogue Corpus : A Data Set for Time-Offset Interactions and Unstructured Dialogue Systems
Alberto Chierici | Nizar Habash | Margarita Bicec

Time-Offset Interaction Applications (TOIAs) are systems that simulate face-to-face conversations between humans and digital human avatars recorded in the past. Developing a well-functioning TOIA involves several research areas : artificial intelligence, human-computer interaction, natural language processing, question answering, and dialogue systems. The first challenges are to define a sensible methodology for data collection and to create useful data sets for training the system to retrieve the best answer to a user’s question. In this paper, we present three main contributions : a methodology for creating the knowledge base for a TOIA, a dialogue corpus, and baselines for single-turn answer retrieval. We develop the methodology using a two-step strategy. First, we let the avatar maker list pairs by intuition, guessing what possible questions a user may ask to the avatar. Second, we record actual dialogues between random individuals and the avatar-maker. We make the Margarita Dialogue Corpus available to the research community. This corpus comprises the knowledge base in text format, the video clips for each answer, and the annotated dialogues.

pdf bib
How Users React to Proactive Voice Assistant Behavior While Driving
Maria Schmidt | Wolfgang Minker | Steffen Werner

Nowadays Personal Assistants (PAs) are available in multiple environments and become increasingly popular to use via voice. Therefore, we aim to provide proactive PA suggestions to car drivers via speech. These suggestions should be neither obtrusive nor increase the drivers’ cognitive load, while enhancing user experience. To assess these factors, we conducted a usability study in which 42 participants perceive proactive voice output in a Wizard-of-Oz study in a driving simulator. Traffic density was varied during a highway drive and it included six in-car-specific use cases. The latter were presented by a proactive voice assistant and in a non-proactive control condition. We assessed the users’ subjective cognitive load and their satisfaction in different questionnaires during the interaction with both PA variants. Furthermore, we analyze the user reactions : both regarding their content and the elapsed response times to PA actions. The results show that proactive assistant behavior is rated similarly positive as non-proactive behavior. Furthermore, the participants agreed to 73.8 % of proactive suggestions. In line with previous research, driving-relevant use cases receive the best ratings, here we reach 82.5 % acceptance. Finally, the users reacted significantly faster to proactive PA actions, which we interpret as less cognitive load compared to non-proactive behavior.

pdf bib
Emotional Speech Corpus for Persuasive Dialogue System
Sara Asai | Koichiro Yoshino | Seitaro Shinagawa | Sakriani Sakti | Satoshi Nakamura

Expressing emotion is known as an efficient way to persuade one’s dialogue partner to accept one’s claim or proposal. Emotional expression in speech can express the speaker’s emotion more directly than using only emotion expression in the text, which will lead to a more persuasive dialogue. In this paper, we built a speech dialogue corpus in a persuasive scenario that uses emotional expressions to build a persuasive dialogue system with emotional expressions. We extended an existing text dialogue corpus by adding variations of emotional responses to cover different combinations of broad dialogue context and a variety of emotional states by crowd-sourcing. Then, we recorded emotional speech consisting of of collected emotional expressions spoken by a voice actor. The experimental results indicate that the collected emotional expressions with their speeches have higher emotional expressiveness for expressing the system’s emotion to users.

pdf bib
Evaluation of Argument Search Approaches in the Context of Argumentative Dialogue Systems
Niklas Rach | Yuki Matsuda | Johannes Daxenberger | Stefan Ultes | Keiichi Yasumoto | Wolfgang Minker

We present an approach to evaluate argument search techniques in view of their use in argumentative dialogue systems by assessing quality aspects of the retrieved arguments. To this end, we introduce a dialogue system that presents arguments by means of a virtual avatar and synthetic speech to users and allows them to rate the presented content in four different categories (Interesting, Convincing, Comprehensible, Relation). The approach is applied in a user study in order to compare two state of the art argument search engines to each other and with a system based on traditional web search. The results show a significant advantage of the two search engines over the baseline. Moreover, the two search engines show significant advantages over each other in different categories, thereby reflecting strengths and weaknesses of the different underlying techniques.

pdf bib
Estimating User Communication Styles for Spoken Dialogue Systems
Juliana Miehle | Isabel Feustel | Julia Hornauer | Wolfgang Minker | Stefan Ultes

We present a neural network approach to estimate the communication style of spoken interaction, namely the stylistic variations elaborateness and directness, and investigate which type of input features to the estimator are necessary to achive good performance. First, we describe our annotated corpus of recordings in the health care domain and analyse the corpus statistics in terms of agreement, correlation and reliability of the ratings. We use this corpus to estimate the elaborateness and the directness of each utterance. We test different feature sets consisting of dialogue act features, grammatical features and linguistic features as input for our classifier and perform classification in two and three classes. Our classifiers use only features that can be automatically derived during an ongoing interaction in any spoken dialogue system without any prior annotation. Our results show that the elaborateness can be classified by only using the dialogue act and the amount of words contained in the corresponding utterance. The directness is a more difficult classification task and additional linguistic features in form of word embeddings improve the classification results. Afterwards, we run a comparison with a support vector machine and a recurrent neural network classifier.

pdf bib
The AICO Multimodal Corpus Data Collection and Preliminary AnalysesAICO Multimodal Corpus – Data Collection and Preliminary Analyses
Kristiina Jokinen

This paper describes data collection and the first explorative research on the AICO Multimodal Corpus. The corpus contains eye-gaze, Kinect, and video recordings of human-robot and human-human interactions, and was collected to study cooperation, engagement and attention of human participants in task-based as well as in chatty type interactive situations. In particular, the goal was to enable comparison between human-human and human-robot interactions, besides studying multimodal behaviour and attention in the different dialogue activities. The robot partner was a humanoid Nao robot, and it was expected that its agent-like behaviour would render humanrobot interactions similar to human-human interaction but also high-light important differences due to the robot’s limited conversational capabilities. The paper reports on the preliminary studies on the corpus, concerning the participants’ eye-gaze and gesturing behaviours, which were chosen as objective measures to study differences in their multimodal behaviour patterns with a human and a robot partner.

pdf bib
RDG-Map : A Multimodal Corpus of Pedagogical Human-Agent Spoken Interactions.RDG-Map: A Multimodal Corpus of Pedagogical Human-Agent Spoken Interactions.
Maike Paetzel | Deepthi Karkada | Ramesh Manuvinakurike

This paper presents a multimodal corpus of 209 spoken game dialogues between a human and a remote-controlled artificial agent. The interactions involve people collaborating with the agent to identify countries on the world map as quickly as possible, which allows studying rapid and spontaneous dialogue with complex anaphoras, disfluent utterances and incorrect descriptions. The corpus consists of two parts : 8 hours of game interactions have been collected with a virtual unembodied agent online and 26.8 hours have been recorded with a physically embodied robot in a research lab. In addition to spoken audio recordings available for both parts, camera recordings and skeleton-, facial expression- and eye-gaze tracking data have been collected for the lab-based part of the corpus. In this paper, we introduce the pedagogical reference resolution game (RDG-Map) and the characteristics of the corpus collected. We also present an annotation scheme we developed in order to study the dialogue strategies utilized by the players. Based on a subset of 330 minutes of interactions annotated so far, we discuss initial insights into these strategies as well as the potential of the corpus for future research.

pdf bib
Alexa in the wild Collecting Unconstrained Conversations with a Modern Voice Assistant in a Public EnvironmentAlexa in the wild” – Collecting Unconstrained Conversations with a Modern Voice Assistant in a Public Environment
Ingo Siegert

Datasets featuring modern voice assistants such as Alexa, Siri, Cortana and others allow an easy study of human-machine interactions. But data collections offering an unconstrained, unscripted public interaction are quite rare. Many studies so far have focused on private usage, short pre-defined task or specific domains. This contribution presents a dataset providing a large amount of unconstrained public interactions with a voice assistant. Up to now around 40 hours of device directed utterances were collected during a science exhibition touring through Germany. The data recording was part of an exhibit that engages visitors to interact with a commercial voice assistant system (Amazon’s ALEXA), but did not restrict them to a specific topic. A specifically developed quiz was starting point of the conversation, as the voice assistant was presented to the visitors as a possible joker for the quiz. But the visitors were not forced to solve the quiz with the help of the voice assistant and thus many visitors had an open conversation. The provided dataset Voice Assistant Conversations in the wild (VACW) includes the transcripts of both visitors requests and Alexa answers, identified topics and sessions as well as acoustic characteristics automatically extractable from the visitors’ audio files.

pdf bib
Dialogue Act Annotation in a Multimodal Corpus of First Encounter Dialogues
Costanza Navarretta | Patrizia Paggio

This paper deals with the annotation of dialogue acts in a multimodal corpus of first encounter dialogues, i.e. face-to- face dialogues in which two people who meet for the first time talk with no particular purpose other than just talking. More specifically, we describe the method used to annotate dialogue acts in the corpus, including the evaluation of the annotations. Then, we present descriptive statistics of the annotation, particularly focusing on which dialogue acts often follow each other across speakers and which dialogue acts overlap with gestural behaviour. Finally, we discuss how feedback is expressed in the corpus by means of feedback dialogue acts with or without co-occurring gestural behaviour, i.e. multimodal vs. unimodal feedback.

pdf bib
A Conversation-Analytic Annotation of Turn-Taking Behavior in Japanese Multi-Party Conversation and its Preliminary AnalysisJapanese Multi-Party Conversation and its Preliminary Analysis
Mika Enomoto | Yasuharu Den | Yuichi Ishimoto

In this study, we propose a conversation-analytic annotation scheme for turn-taking behavior in multi-party conversations. The annotation scheme is motivated by a proposal of a proper model of turn-taking incorporating various ideas developed in the literature of conversation analysis. Our annotation consists of two sets of tags : the beginning and the ending type of the utterance. Focusing on the ending-type tags, in some cases combined with the beginning-type tags, we emphasize the importance of the distinction among four selection types : i) selecting other participant as next speaker, ii) not selecting next speaker but followed by a switch of the speakership, iii) not selecting next speaker and followed by a continuation of the speakership, and iv)being inside a multi-unit turn. Based on the annotation of Japanese multi-party conversations, we analyze how syntactic and prosodic features of utterances vary across the four selection types. The results show that the above four-way distinction is essential to account for the distributions of the syntactic and prosodic features, suggesting the insufficiency of previous turn-taking models that do not consider the distinction between i) and ii) or between ii) or iii).

pdf bib
Dialogue-AMR : Abstract Meaning Representation for DialogueAMR: Abstract Meaning Representation for Dialogue
Claire Bonial | Lucia Donatelli | Mitchell Abrams | Stephanie M. Lukin | Stephen Tratz | Matthew Marge | Ron Artstein | David Traum | Clare Voss

This paper describes a schema that enriches Abstract Meaning Representation (AMR) in order to provide a semantic representation for facilitating Natural Language Understanding (NLU) in dialogue systems. AMR offers a valuable level of abstraction of the propositional content of an utterance ; however, it does not capture the illocutionary force or speaker’s intended contribution in the broader dialogue context (e.g., make a request or ask a question), nor does it capture tense or aspect. We explore dialogue in the domain of human-robot interaction, where a conversational robot is engaged in search and navigation tasks with a human partner. To address the limitations of standard AMR, we develop an inventory of speech acts suitable for our domain, and present Dialogue-AMR, an enhanced AMR that represents not only the content of an utterance, but the illocutionary force behind it, as well as tense and aspect. To showcase the coverage of the schema, we use both manual and automatic methods to construct the DialAMR corpusa corpus of human-robot dialogue annotated with standard AMR and our enriched Dialogue-AMR schema. Our automated methods can be used to incorporate AMR into a larger NLU pipeline supporting human-robot dialogue.

pdf bib
Relation between Degree of Empathy for Narrative Speech and Type of Responsive Utterance in Attentive Listening
Koichiro Ito | Masaki Murata | Tomohiro Ohno | Shigeki Matsubara

Nowadays, spoken dialogue agents such as communication robots and smart speakers listen to narratives of humans. In order for such an agent to be recognized as a listener of narratives and convey the attitude of attentive listening, it is necessary to generate responsive utterances. Moreover, responsive utterances can express empathy to narratives and showing an appropriate degree of empathy to narratives is significant for enhancing speaker’s motivation. The degree of empathy shown by responsive utterances is thought to depend on their type. However, the relation between responsive utterances and degrees of the empathy has not been explored yet. This paper describes the classification of responsive utterances based on the degree of empathy in order to explain that relation. In this research, responsive utterances are classified into five levels based on the effect of utterances and literature on attentive listening. Quantitative evaluations using 37,995 responsive utterances showed the appropriateness of the proposed classification.

pdf bib
Predicting Ratings of Real Dialogue Participants from Artificial Data and Ratings of Human Dialogue Observers
Kallirroi Georgila | Carla Gordon | Volodymyr Yanov | David Traum

We collected a corpus of dialogues in a Wizard of Oz (WOz) setting in the Internet of Things (IoT) domain. We asked users participating in these dialogues to rate the system on a number of aspects, namely, intelligence, naturalness, personality, friendliness, their enjoyment, overall quality, and whether they would recommend the system to others. Then we asked dialogue observers, i.e., Amazon Mechanical Turkers (MTurkers), to rate these dialogues on the same aspects. We also generated simulated dialogues between dialogue policies and simulated users and asked MTurkers to rate them again on the same aspects. Using linear regression, we developed dialogue evaluation functions based on features from the simulated dialogues and the MTurkers’ ratings, the WOz dialogues and the MTurkers’ ratings, and the WOz dialogues and the WOz participants’ ratings. We applied all these dialogue evaluation functions to a held-out portion of our WOz dialogues, and we report results on the predictive power of these different types of dialogue evaluation functions. Our results suggest that for three conversational aspects (intelligence, naturalness, overall quality) just training evaluation functions on simulated data could be sufficient.

pdf bib
Which Model Should We Use for a Real-World Conversational Dialogue System? a Cross-Language Relevance Model or a Deep Neural Net?
Seyed Hossein Alavi | Anton Leuski | David Traum

We compare two models for corpus-based selection of dialogue responses : one based on cross-language relevance with a cross-language LSTM model. Each model is tested on multiple corpora, collected from two different types of dialogue source material. Results show that while the LSTM model performs adequately on a very large corpus (millions of utterances), its performance is dominated by the cross-language relevance model for a more moderate-sized corpus (ten thousands of utterances).

pdf bib
AMUSED : A Multi-Stream Vector Representation Method for Use in Natural DialogueAMUSED: A Multi-Stream Vector Representation Method for Use in Natural Dialogue
Gaurav Kumar | Rishabh Joshi | Jaspreet Singh | Promod Yenigalla

The problem of building a coherent and non-monotonous conversational agent with proper discourse and coverage is still an area of open research. Current architectures only take care of semantic and contextual information for a given query and fail to completely account for syntactic and external knowledge which are crucial for generating responses in a chit-chat system. To overcome this problem, we propose an end to end multi-stream deep learning architecture that learns unified embeddings for query-response pairs by leveraging contextual information from memory networks and syntactic information by incorporating Graph Convolution Networks (GCN) over their dependency parse. A stream of this network also utilizes transfer learning by pre-training a bidirectional transformer to extract semantic representation for each input sentence and incorporates external knowledge through the neighborhood of the entities from a Knowledge Base (KB). We benchmark these embeddings on the next sentence prediction task and significantly improve upon the existing techniques. Furthermore, we use AMUSED to represent query and responses along with its context to develop a retrieval based conversational agent which has been validated by expert linguists to have comprehensive engagement with humans.

pdf bib
An Annotation Approach for Social and Referential Gaze in Dialogue
Vidya Somashekarappa | Christine Howes | Asad Sayeed

This paper introduces an approach for annotating eye gaze considering both its social and the referential functions in multi-modal human-human dialogue. Detecting and interpreting the temporal patterns of gaze behavior cues is natural for humans and also mostly an unconscious process. However, these cues are difficult for conversational agents such as robots or avatars to process or generate. The key factor is to recognize these variants and carry out a successful conversation, as misinterpretation can lead to total failure of the given interaction. This paper introduces an annotation scheme for eye-gaze in human-human dyadic interactions that is intended to facilitate the learning of eye-gaze patterns in multi-modal natural dialogue.

pdf bib
The Royal Society Corpus 6.0 : Providing 300 + Years of Scientific Writing for Humanistic Study
Stefan Fischer | Jörg Knappen | Katrin Menzel | Elke Teich

We present a new, extended version of the Royal Society Corpus (RSC), a diachronic corpus of scientific English now covering 300 + years of scientific writing (16651996). The corpus comprises 47 837 texts, primarily scientific articles, and is based on publications of the Royal Society of London, mainly its Philosophical Transactions and Proceedings. The corpus has been built on the basis of the FAIR principles and is freely available under a Creative Commons license, excluding copy-righted parts. We provide information on how the corpus can be found, the file formats available for download as well as accessibility via a web-based corpus query platform. We show a number of analytic tools that we have implemented for better usability and provide an example of use of the corpus for linguistic analysis as well as examples of subsequent, external uses of earlier releases. We place the RSC against the background of existing English diachronic / scientific corpora, elaborating on its value for linguistic and humanistic study.

pdf bib
RiQuA : A Corpus of Rich Quotation Annotation for English Literary TextRiQuA: A Corpus of Rich Quotation Annotation for English Literary Text
Sean Papay | Sebastian Padó

We introduce RiQuA (RIch QUotation Annotations), a corpus that provides quotations, including their interpersonal structure (speakers and addressees) for English literary text. The corpus comprises 11 works of 19th-century literature that were manually doubly annotated for direct and indirect quotations. For each quotation, its span, speaker, addressee, and cue are identified (if present). This provides a rich view of dialogue structures not available from other available corpora. We detail the process of creating this dataset, discuss the annotation guidelines, and analyze the resulting corpus in terms of inter-annotator agreement and its properties. RiQuA, along with its annotations guidelines and associated scripts, are publicly available for use, modification, and experimentation.

pdf bib
The BDCames Collection of Portuguese Literary Documents : a Research Resource for Digital Humanities and Language TechnologyBDCamões Collection of Portuguese Literary Documents: a Research Resource for Digital Humanities and Language Technology
Sara Grilo | Márcia Bolrinha | João Silva | Rui Vaz | António Branco

This paper presents the BDCames Collection of Portuguese Literary Documents, a new corpus of literary texts written in Portuguese that in its inaugural version includes close to 4 million words from over 200 complete documents from 83 authors in 14 genres, covering a time span from the 16th to the 21st century, and adhering to different orthographic conventions. Many of the texts in the corpus have also been automatically parsed with state-of-the-art language processing tools, forming the BDCames Treebank subcorpus. This set of characteristics makes of BDCames an invaluable resource for research in language technology (e.g. authorship detection, genre classification, etc.) and in language science and digital humanities (e.g. comparative literature, diachronic linguistics, etc.

pdf bib
NLP Scholar : A Dataset for Examining the State of NLP ResearchNLP Scholar: A Dataset for Examining the State of NLP Research
Saif M. Mohammad

Google Scholar is the largest web search engine for academic literature that also provides access to rich metadata associated with the papers. The ACL Anthology (AA) is the largest repository of articles on Natural Language Processing (NLP). We extracted information from AA for about 44 thousand NLP papers and identified authors who published at least three papers there. We then extracted citation information from Google Scholar for all their papers (not just their AA papers). This resulted in a dataset of 1.1 million papers and associated Google Scholar information. We aligned the information in the AA and Google Scholar datasets to create the NLP Scholar Dataset a single unified source of information (from both AA and Google Scholar) for tens of thousands of NLP papers. It can be used to identify broad trends in productivity, focus, and impact of NLP research. We present here initial work on analyzing the volume of research in NLP over the years and identifying the most cited papers in NLP. We also list a number of additional potential applications.

pdf bib
The DReaM Corpus : A Multilingual Annotated Corpus of Grammars for the World’s LanguagesDReaM Corpus: A Multilingual Annotated Corpus of Grammars for the World’s Languages
Shafqat Mumtaz Virk | Harald Hammarström | Markus Forsberg | Søren Wichmann

There exist as many as 7000 natural languages in the world, and a huge number of documents describing those languages have been produced over the years. Most of those documents are in paper format. Any attempts to use modern computational techniques and tools to process those documents will require them to be digitized first. In this paper, we report a multilingual digitized version of thousands of such documents searchable through some well-established corpus infrastructures. The corpus is annotated with various meta, word, and text level attributes to make searching and analysis easier and more useful.

pdf bib
Identification of Indigenous Knowledge Concepts through Semantic Networks, Spelling Tools and Word Embeddings
Renato Rocha Souza | Amelie Dorn | Barbara Piringer | Eveline Wandl-Vogt

In order to access indigenous, regional knowledge contained in language corpora, semantic tools and network methods are most typically employed. In this paper we present an approach for the identification of dialectal variations of words, or words that do not pertain to High German, on the example of non-standard language legacy collection questionnaires of the Bavarian Dialects in Austria (DB). Based on selected cultural categories relevant to the wider project context, common words from each of these cultural categories and their lemmas using GermaLemma were identified. Through word embedding models the semantic vicinity of each word was explored, followed by the use of German Wordnet (Germanet) and the Hunspell tool. Whilst none of these tools have a comprehensive coverage of standard German words, they serve as an indication of dialects in specific semantic hierarchies. Methods and tools applied in this study may serve as an example for other similar projects dealing with non-standard or endangered language collections, aiming to access, analyze and ultimately preserve native regional language heritage.

pdf bib
An Annotated Corpus of Adjective-Adverb Interfaces in Romance LanguagesRomance Languages
Katharina Gerhalter | Gerlinde Schneider | Christopher Pollin | Martin Hummel

The final outcome of the project Open Access Database : Adjective-Adverb Interfaces in Romance is an annotated and lemmatised corpus of various linguistic phenomena related to Romance adjectives with adverbial functions. The data is published under open-access and aims to serve linguistic research based on transparent and accessible corpus-based data. The annotation model was developed to offer a cross-linguistic categorization model for the heterogeneous word-class adverb, based on its diverse forms, functions and meanings. The project focuses on the interoperability and accessibility of data, with particular respect to reusability in the sense of the FAIR Data Principles. Topics presented by this paper include data compilation and creation, annotation in XML / TEI, data preservation and publication process by means of the GAMS repository and accessibility via a search interface. These aspects are tied together by semantic technologies, using an ontology-based approach.

pdf bib
Chinese Discourse Parsing : Model and EvaluationChinese Discourse Parsing: Model and Evaluation
Lin Chuan-An | Shyh-Shiun Hung | Hen-Hsen Huang | Hsin-Hsi Chen

Chinese discourse parsing, which aims to identify the hierarchical relationships of Chinese elementary discourse units, has not yet a consistent evaluation metric. Although Parseval is commonly used, variations of evaluation differ from three aspects : micro vs. macro F1 scores, binary vs. multiway ground truth, and left-heavy vs. right-heavy binarization. In this paper, we first propose a neural network model that unifies a pre-trained transformer and CKY-like algorithm, and then compare it with the previous models with different evaluation scenarios. The experimental results show that our model outperforms the previous systems. We conclude that (1) the pre-trained context embedding provides effective solutions to deal with implicit semantics in Chinese texts, and (2) using multiway ground truth is helpful since different binarization approaches lead to significant differences in performance.

pdf bib
The Potsdam Commentary Corpus 2.2 : Extending Annotations for Shallow Discourse ParsingPotsdam Commentary Corpus 2.2: Extending Annotations for Shallow Discourse Parsing
Peter Bourgonje | Manfred Stede

We present the Potsdam Commentary Corpus 2.2, a German corpus of news editorials annotated on several different levels. New in the 2.2 version of the corpus are two additional annotation layers for coherence relations following the Penn Discourse TreeBank framework. Specifically, we add relation senses to an already existing layer of discourse connectives and their arguments, and we introduce a new layer with additional coherence relation types, resulting in a German corpus that mirrors the PDTB. The aim of this is to increase usability of the corpus for the task of shallow discourse parsing. In this paper, we provide inter-annotator agreement figures for the new annotations and compare corpus statistics based on the new annotations to the equivalent statistics extracted from the PDTB.

pdf bib
DiMLex-Bangla : A Lexicon of Bangla Discourse ConnectivesDiMLex-Bangla: A Lexicon of Bangla Discourse Connectives
Debopam Das | Manfred Stede | Soumya Sankar Ghosh | Lahari Chatterjee

We present DiMLex-Bangla, a newly developed lexicon of discourse connectives in Bangla. The lexicon, upon completion of its first version, contains 123 Bangla connective entries, which are primarily compiled from the linguistic literature and translation of English discourse connectives. The lexicon compilation is later augmented by adding more connectives from a currently developed corpus, called the Bangla RST Discourse Treebank (Das and Stede, 2018). DiMLex-Bangla provides information on syntactic categories of Bangla connectives, their discourse semantics and non-connective uses (if any). It uses the format of the German connective lexicon DiMLex (Stede and Umbach, 1998), which provides a cross-linguistically applicable XML schema. The resource is the first of its kind in Bangla, and is freely available for use in studies on discourse structure and computational applications.

pdf bib
What Speakers really Mean when they Ask Questions : Classification of Intentions with a Supervised Approach
Angèle Barbedette | Iris Eshkol-Taravella

This paper focuses on the automatic detection of hidden intentions of speakers in questions asked during meals. Our corpus is composed of a set of transcripts of spontaneous oral conversations from ESLO’s corpora. We suggest a typology of these intentions based on our research work and the exploration and annotation of the corpus, in which we define two explicit categories (request for agreement and request for information) and three implicit categories (opinion, will and doubt). We implement a supervised automatic classification model based on annotated data and selected linguistic features and we evaluate its results and performances. We finally try to interpret these results by looking more deeply and specifically into the predictions of the algorithm and the features it used. There are many motivations for this work which are part of ongoing challenges such as opinion analysis, irony detection or the development of conversational agents.

pdf bib
Modeling Dialogue in Conversational Cognitive Health Screening Interviews
Shahla Farzana | Mina Valizadeh | Natalie Parde

Automating straightforward clinical tasks can reduce workload for healthcare professionals, increase accessibility for geographically-isolated patients, and alleviate some of the economic burdens associated with healthcare. A variety of preliminary screening procedures are potentially suitable for automation, and one such domain that has remained underexplored to date is that of structured clinical interviews. A task-specific dialogue agent is needed to automate the collection of conversational speech for further (either manual or automated) analysis, and to build such an agent, a dialogue manager must be trained to respond to patient utterances in a manner similar to a human interviewer. To facilitate the development of such an agent, we propose an annotation schema for assigning dialogue act labels to utterances in patient-interviewer conversations collected as part of a clinically-validated cognitive health screening task. We build a labeled corpus using the schema, and show that it is characterized by high inter-annotator agreement. We establish a benchmark dialogue act classification model for the corpus, thereby providing a proof of concept for the proposed annotation schema. The resulting dialogue act corpus is the first such corpus specifically designed to facilitate automated cognitive health screening, and lays the groundwork for future exploration in this area.

pdf bib
Stigma Annotation Scheme and Stigmatized Language Detection in Health-Care Discussions on Social Media
Nadiya Straton | Hyeju Jang | Raymond Ng

Much research has been done within the social sciences on the interpretation and influence of stigma on human behaviour and health, which result in out-of-group exclusion, distancing, cognitive separation, status loss, discrimination, in-group pressure, and often lead to disengagement, non-adherence to treatment plan, and prescriptions by the doctor. However, little work has been conducted on computational identification of stigma in general and in social media discourse in particular. In this paper, we develop the annotation scheme and improve the annotation process for stigma identification, which can be applied to other health-care domains. The data from pro-vaccination and anti-vaccination discussion groups are annotated by trained annotators who have professional background in social science and health-care studies, therefore the group can be considered experts on the subject in comparison to non-expert crowd. Amazon MTurk annotators is another group of annotator with no knowledge on their education background, they are initially treated as non-expert crowd on the subject matter of stigma. We analyze the annotations with visualisation techniques, features from LIWC (Linguistic Inquiry and Word Count) list and make prediction based on bi-grams with traditional and deep learning models. Data augmentation method and application of CNN show high performance accuracy in comparison to other models. Success of the rigorous annotation process on identifying stigma is reconfirmed by achieving high prediction rate with CNN.

pdf bib
Discovering Biased News Articles Leveraging Multiple Human Annotations
Konstantina Lazaridou | Alexander Löser | Maria Mestre | Felix Naumann

Unbiased and fair reporting is an integral part of ethical journalism. Yet, political propaganda and one-sided views can be found in the news and can cause distrust in media. Both accidental and deliberate political bias affect the readers and shape their views. We contribute to a trustworthy media ecosystem by automatically identifying politically biased news articles. We introduce novel corpora annotated by two communities, i.e., domain experts and crowd workers, and we also consider automatic article labels inferred by the newspapers’ ideologies. Our goal is to compare domain experts to crowd workers and also to prove that media bias can be detected automatically. We classify news articles with a neural network and we also improve our performance in a self-supervised manner.

pdf bib
Corpora and Baselines for Humour Recognition in PortuguesePortuguese
Hugo Gonçalo Oliveira | André Clemêncio | Ana Alves

Having in mind the lack of work on the automatic recognition of verbal humour in Portuguese, a topic connected with fluency in a natural language, we describe the creation of three corpora, covering two styles of humour and four sources of non-humorous text, that may be used for related studies. We then report on some experiments where the created corpora were used for training and testing computational models that exploit content and linguistic features for humour recognition. The obtained results helped us taking some conclusions about this challenge and may be seen as baselines for those willing to tackle it in the future, using the same corpora.

pdf bib
Using Deep Neural Networks with Intra- and Inter-Sentence Context to Classify Suicidal Behaviour
Xingyi Song | Johnny Downs | Sumithra Velupillai | Rachel Holden | Maxim Kikoler | Kalina Bontcheva | Rina Dutta | Angus Roberts

Identifying statements related to suicidal behaviour in psychiatric electronic health records (EHRs) is an important step when modeling that behaviour, and when assessing suicide risk. We apply a deep neural network based classification model with a lightweight context encoder, to classify sentence level suicidal behaviour in EHRs. We show that incorporating information from sentences to left and right of the target sentence significantly improves classification accuracy. Our approach achieved the best performance when classifying suicidal behaviour in Autism Spectrum Disorder patient records. The results could have implications for suicidality research and clinical surveillance.

pdf bib
Habibi-a multi Dialect multi National Arabic Song Lyrics CorpusArabic Song Lyrics Corpus
Mahmoud El-Haj

This paper introduces Habibi the first Arabic Song Lyrics corpus. The corpus comprises more than 30,000 Arabic song lyrics in 6 Arabic dialects for singers from 18 different Arabic countries. The lyrics are segmented into more than 500,000 sentences (song verses) with more than 3.5 million words. I provide the corpus in both comma separated value (csv) and annotated plain text (txt) file formats. In addition, I converted the csv version into JavaScript Object Notation (json) and eXtensible Markup Language (xml) file formats. To experiment with the corpus I run extensive binary and multi-class experiments for dialect and country-of-origin identification. The identification tasks include the use of several classical machine learning and deep learning models utilising different word embeddings. For the binary dialect identification task the best performing classifier achieved a testing accuracy of 93 %. This was achieved using a word-based Convolutional Neural Network (CNN) utilising a Continuous Bag of Words (CBOW) word embeddings model. The results overall show all classical and deep learning models to outperform our baseline, which demonstrates the suitability of the corpus for both dialect and country-of-origin identification tasks. I am making the corpus and the trained CBOW word embeddings freely available for research purposes.

pdf bib
Email Classification Incorporating Social Networks and Thread Structure
Sakhar Alkhereyf | Owen Rambow

Existing methods for different document classification tasks in the context of social networks typically only capture the semantics of texts, while ignoring the users who exchange the text and the network they form. However, some work has shown that incorporating the social network information in addition to information from language is effective for various NLP applications including sentiment analysis, inferring user attributes, and predicting inter-personal relations. In this paper, we present an empirical study of email classification into Business and Personal categories. We represent the email communication using various graph structures. As features, we use both the textual information from the email content and social network information from the communication graphs. We also model the thread structure for emails. We focus on detecting personal emails, and we evaluate our methods on two corpora, only one of which we train on. The experimental results reveal that incorporating social network information improves over the performance of an approach based on textual information only. The results also show that considering the thread structure of emails improves the performance further. Furthermore, our approach improves over a state-of-the-art baseline which uses node embeddings based on both lexical and social network information.

pdf bib
Alector : A Parallel Corpus of Simplified French Texts with Alignments of Misreadings by Poor and Dyslexic ReadersAlector: A Parallel Corpus of Simplified French Texts with Alignments of Misreadings by Poor and Dyslexic Readers
Núria Gala | Anaïs Tack | Ludivine Javourey-Drevet | Thomas François | Johannes C. Ziegler

In this paper, we present a new parallel corpus addressed to researchers, teachers, and speech therapists interested in text simplification as a means of alleviating difficulties in children learning to read. The corpus is composed of excerpts drawn from 79 authentic literary (tales, stories) and scientific (documentary) texts commonly used in French schools for children aged between 7 to 9 years old. The excerpts were manually simplified at the lexical, morpho-syntactic, and discourse levels in order to propose a parallel corpus for reading tests and for the development of automatic text simplification tools. A sample of 21 poor-reading and dyslexic children with an average reading delay of 2.5 years read a portion of the corpus. The transcripts of readings errors were integrated into the corpus with the goal of identifying lexical difficulty in the target population. By means of statistical testing, we provide evidence that the manual simplifications significantly reduced reading errors, highlighting that the words targeted for simplification were not only well-chosen but also substituted with substantially easier alternatives. The entire corpus is available for consultation through a web interface and available on demand for research purposes.

pdf bib
An Evaluation of Progressive Neural Networksfor Transfer Learning in Natural Language Processing
Abdul Moeed | Gerhard Hagerer | Sumit Dugar | Sarthak Gupta | Mainak Ghosh | Hannah Danner | Oliver Mitevski | Andreas Nawroth | Georg Groh

A major challenge in modern neural networks is the utilization of previous knowledge for new tasks in an effective manner, otherwise known as transfer learning. Fine-tuning, the most widely used method for achieving this, suffers from catastrophic forgetting. The problem is often exacerbated in natural language processing (NLP). In this work, we assess progressive neural networks (PNNs) as an alternative to fine-tuning. The evaluation is based on common NLP tasks such as sequence labeling and text classification. By gauging PNNs across a range of architectures, datasets, and tasks, we observe improvements over the baselines throughout all experiments.

pdf bib
DecOp : A Multilingual and Multi-domain Corpus For Detecting Deception In Typed TextDecOp: A Multilingual and Multi-domain Corpus For Detecting Deception In Typed Text
Pasquale Capuozzo | Ivano Lauriola | Carlo Strapparava | Fabio Aiolli | Giuseppe Sartori

In recent years, the increasing interest in the development of automatic approaches for unmasking deception in online sources led to promising results. Nonetheless, among the others, two major issues remain still unsolved : the stability of classifiers performances across different domains and languages. Tackling these issues is challenging since labelled corpora involving multiple domains and compiled in more than one language are few in the scientific literature. For filling this gap, in this paper we introduce DecOp (Deceptive Opinions), a new language resource developed for automatic deception detection in cross-domain and cross-language scenarios. DecOp is composed of 5000 examples of both truthful and deceitful first-person opinions balanced both across five different domains and two languages and, to the best of our knowledge, is the largest corpus allowing cross-domain and cross-language comparisons in deceit detection tasks. In this paper, we describe the collection procedure of the DecOp corpus and his main characteristics. Moreover, the human performance on the DecOp test-set and preliminary experiments by means of machine learning models based on Transformer architecture are shown.

pdf bib
VICTOR : a Dataset for Brazilian Legal Documents ClassificationVICTOR: a Dataset for Brazilian Legal Documents Classification
Pedro Henrique Luz de Araujo | Teófilo Emídio de Campos | Fabricio Ataides Braz | Nilton Correia da Silva

This paper describes VICTOR, a novel dataset built from Brazil’s Supreme Court digitalized legal documents, composed of more than 45 thousand appeals, which includes roughly 692 thousand documentsabout 4.6 million pages. The dataset contains labeled text data and supports two types of tasks : document type classification ; and theme assignment, a multilabel problem. We present baseline results using bag-of-words models, convolutional neural networks, recurrent neural networks and boosting algorithms. We also experiment using linear-chain Conditional Random Fields to leverage the sequential nature of the lawsuits, which we find to lead to improvements on document type classification. Finally we compare a theme classification approach where we use domain knowledge to filter out the less informative document pages to the default one where we use all pages. Contrary to the Court experts’ expectations, we find that using all available data is the better method. We make the dataset available in three versions of different sizes and contents to encourage explorations of better models and techniques.

pdf bib
Abusive language in Spanish children and young teenager’s conversations : data preparation and short text classification with contextual word embeddingsSpanish children and young teenager’s conversations: data preparation and short text classification with contextual word embeddings
Marta R. Costa-jussà | Esther González | Asuncion Moreno | Eudald Cumalat

Abusive texts are reaching the interests of the scientific and social community. How to automatically detect them is onequestion that is gaining interest in the natural language processing community. The main contribution of this paper is toevaluate the quality of the recently developed Spanish Database for cyberbullying prevention for the purpose of trainingclassifiers on detecting abusive short texts. We compare classical machine learning techniques to the use of a more ad-vanced model : the contextual word embeddings in the particular case of classification of abusive short-texts for the Spanishlanguage. As contextual word embeddings, we use Bidirectional Encoder Representation from Transformers (BERT), pro-posed at the end of 2018. We show that BERT mostly outperforms classical techniques. Far beyond the experimentalimpact of our research, this project aims at planting the seeds for an innovative technological tool with a high potentialsocial impact and aiming at being part of the initiatives in artificial intelligence for social good.

pdf bib
SOLO : A Corpus of Tweets for Examining the State of Being AloneSOLO: A Corpus of Tweets for Examining the State of Being Alone
Svetlana Kiritchenko | Will Hipson | Robert Coplan | Saif M. Mohammad

The state of being alone can have a substantial impact on our lives, though experiences with time alone diverge significantly among individuals. Psychologists distinguish between the concept of solitude, a positive state of voluntary aloneness, and the concept of loneliness, a negative state of dissatisfaction with the quality of one’s social interactions. Here, for the first time, we conduct a large-scale computational analysis to explore how the terms associated with the state of being alone are used in online language. We present SOLO (State of Being Alone), a corpus of over 4 million tweets collected with query terms solitude, lonely, and loneliness. We use SOLO to analyze the language and emotions associated with the state of being alone. We show that the term solitude tends to co-occur with more positive, high-dominance words (e.g., enjoy, bliss) while the terms lonely and loneliness frequently co-occur with negative, low-dominance words (e.g., scared, depressed), which confirms the conceptual distinctions made in psychology. We also show that women are more likely to report on negative feelings of being lonely as compared to men, and there are more teenagers among the tweeters that use the word lonely than among the tweeters that use the word solitude.

pdf bib
Korean-Specific Emotion Annotation Procedure Using N-Gram-Based Distant Supervision and Korean-Specific-Feature-Based Distant SupervisionKorean-Specific Emotion Annotation Procedure Using N-Gram-Based Distant Supervision and Korean-Specific-Feature-Based Distant Supervision
Young-Jun Lee | Chae-Gyun Lim | Ho-Jin Choi

Detecting emotions from texts is considerably important in an NLP task, but it has the limitation of the scarcity of manually labeled data. To overcome this limitation, many researchers have annotated unlabeled data with certain frequently used annotation procedures. However, most of these studies are focused mainly on English and do not consider the characteristics of the Korean language. In this paper, we present a Korean-specific annotation procedure, which consists of two parts, namely n-gram-based distant supervision and Korean-specific-feature-based distant supervision. We leverage the distant supervision with the n-gram and Korean emotion lexicons. Then, we consider the Korean-specific emotion features. Through experiments, we showed the effectiveness of our procedure by comparing with the KTEA dataset. Additionally, we constructed a large-scale emotion-labeled dataset, Korean Movie Review Emotion (KMRE) Dataset, using our procedure. In order to construct our dataset, we used a large-scale sentiment movie review corpus as the unlabeled dataset. Moreover, we used a Korean emotion lexicon provided by KTEA. We also performed an emotion classification task and a human evaluation on the KMRE dataset.

pdf bib
An Emotional Mess ! Deciding on a Framework for Building a Dutch Emotion-Annotated CorpusDutch Emotion-Annotated Corpus
Luna De Bruyne | Orphee De Clercq | Veronique Hoste

Seeing the myriad of existing emotion models, with the categorical versus dimensional opposition the most important dividing line, building an emotion-annotated corpus requires some well thought-out strategies concerning framework choice. In our work on automatic emotion detection in Dutch texts, we investigate this problem by means of two case studies. We find that the labels joy, love, anger, sadness and fear are well-suited to annotate texts coming from various domains and topics, but that the connotation of the labels strongly depends on the origin of the texts. Moreover, it seems that information is lost when an emotional state is forcedly classified in a limited set of categories, indicating that a bi-representational format is desirable when creating an emotion corpus.

pdf bib
Identification of Primary and Collateral Tracks in Stuttered Speech
Rachid Riad | Anne-Catherine Bachoud-Lévi | Frank Rudzicz | Emmanuel Dupoux

Disfluent speech has been previously addressed from two main perspectives : the clinical perspective focusing on diagnostic, and the Natural Language Processing (NLP) perspective aiming at modeling these events and detect them for downstream tasks. In addition, previous works often used different metrics depending on whether the input features are text or speech, making it difficult to compare the different contributions. Here, we introduce a new evaluation framework for disfluency detection inspired by the clinical and NLP perspective together with the theory of performance from (Clark, 1996) which distinguishes between primary and collateral tracks. We introduce a novel forced-aligned disfluency dataset from a corpus of semi-directed interviews, and present baseline results directly comparing the performance of text-based features (word and span information) and speech-based (acoustic-prosodic information). Finally, we introduce new audio features inspired by the word-based span features. We show experimentally that using these features outperformed the baselines for speech-based predictions on the present dataset.

pdf bib
HardEval : Focusing on Challenging Tokens to Assess Robustness of NERHardEval: Focusing on Challenging Tokens to Assess Robustness of NER
Gabriel Bernier-Colborne | Phillippe Langlais

To assess the robustness of NER systems, we propose an evaluation method that focuses on subsets of tokens that represent specific sources of errors : unknown words and label shift or ambiguity. These subsets provide a system-agnostic basis for evaluating specific sources of NER errors and assessing room for improvement in terms of robustness. We analyze these subsets of challenging tokens in two widely-used NER benchmarks, then exploit them to evaluate NER systems in both in-domain and out-of-domain settings. Results show that these challenging tokens explain the majority of errors made by modern NER systems, although they represent only a small fraction of test tokens. They also indicate that label shift is harder to deal with than unknown words, and that there is much more room for improvement than the standard NER evaluation procedure would suggest. We hope this work will encourage NLP researchers to adopt rigorous and meaningful evaluation methods, and will help them develop more robust models.

pdf bib
Linguistic Appropriateness and Pedagogic Usefulness of Reading Comprehension Questions
Andrea Horbach | Itziar Aldabe | Marie Bexte | Oier Lopez de Lacalle | Montse Maritxalar

Automatic generation of reading comprehension questions is a topic receiving growing interest in the NLP community, but there is currently no consensus on evaluation metrics and many approaches focus on linguistic quality only while ignoring the pedagogic value and appropriateness of questions. This paper overcomes such weaknesses by a new evaluation scheme where questions from the questionnaire are structured in a hierarchical way to avoid confronting human annotators with evaluation measures that do not make sense for a certain question. We show through an annotation study that our scheme can be applied, but that expert annotators with some level of expertise are needed. We also created and evaluated two new evaluation data sets from the biology domain for Basque and German, composed of questions written by people with an educational background, which will be publicly released. Results show that manually generated questions are in general both of higher linguistic as well as pedagogic quality and that among the human generated questions, teacher-generated ones tend to be most useful.

pdf bib
An In-Depth Comparison of 14 Spelling Correction Tools on a Common Benchmark
Markus Näther

Determining and correcting spelling and grammar errors in text is an important but surprisingly difficult task. There are several reasons why this remains challenging. Errors may consist of simple typing errors like deleted, substituted, or wrongly inserted letters, but may also consist of word confusions where a word was replaced by another one. In addition, words may be erroneously split into two parts or get concatenated. Some words can contain hyphens, because they were split at the end of a line or are compound words with a mandatory hyphen. In this paper, we provide an extensive evaluation of 14 spelling correction tools on a common benchmark. In particular, the evaluation provides a detailed comparison with respect to 12 error categories. The benchmark consists of sentences from the English Wikipedia, which were distorted using a realistic error model. Measuring the quality of an algorithm with respect to these error categories requires an alignment of the original text, the distorted text and the corrected text provided by the tool. We make our benchmark generation and evaluation tools publicly available.

pdf bib
Stress Test Evaluation of Transformer-based Models in Natural Language Understanding Tasks
Carlos Aspillaga | Andrés Carvallo | Vladimir Araujo

There has been significant progress in recent years in the field of Natural Language Processing thanks to the introduction of the Transformer architecture. Current state-of-the-art models, via a large number of parameters and pre-training on massive text corpus, have shown impressive results on several downstream tasks. Many researchers have studied previous (non-Transformer) models to understand their actual behavior under different scenarios, showing that these models are taking advantage of clues or failures of datasets and that slight perturbations on the input data can severely reduce their performance. In contrast, recent models have not been systematically tested with adversarial-examples in order to show their robustness under severe stress conditions. For that reason, this work evaluates three Transformer-based models (RoBERTa, XLNet, and BERT) in Natural Language Inference (NLI) and Question Answering (QA) tasks to know if they are more robust or if they have the same flaws as their predecessors. As a result, our experiments reveal that RoBERTa, XLNet and BERT are more robust than recurrent neural network models to stress tests for both NLI and QA tasks. Nevertheless, they are still very fragile and demonstrate various unexpected behaviors, thus revealing that there is still room for future improvement in this field.

pdf bib
Brand-Product Relation Extraction Using Heterogeneous Vector Space Representations
Arkadiusz Janz | Łukasz Kopociński | Maciej Piasecki | Agnieszka Pluwak

Relation Extraction is a fundamental NLP task. In this paper we investigate the impact of underlying text representation on the performance of neural classification models in the task of Brand-Product relation extraction. We also present the methodology of preparing annotated textual corpora for this task and we provide valuable insight into the properties of Brand-Product relations existing in textual corpora. The problem is approached from a practical angle of applications Relation Extraction in facilitating commercial Internet monitoring.

pdf bib
TableBank : Table Benchmark for Image-based Table Detection and RecognitionTableBank: Table Benchmark for Image-based Table Detection and Recognition
Minghao Li | Lei Cui | Shaohan Huang | Furu Wei | Ming Zhou | Zhoujun Li

We present TableBank, a new image-based table detection and recognition dataset built with novel weak supervision from Word and Latex documents on the internet. Existing research for image-based table detection and recognition usually fine-tunes pre-trained models on out-of-domain data with a few thousand human-labeled examples, which is difficult to generalize on real-world applications. With TableBank that contains 417 K high quality labeled tables, we build several strong baselines using state-of-the-art models with deep neural networks. We make TableBank publicly available and hope it will empower more deep learning approaches in the table detection and recognition task. The dataset and models can be downloaded from

pdf bib
WIKIR : A Python Toolkit for Building a Large-scale Wikipedia-based English Information Retrieval DatasetWIKIR: A Python Toolkit for Building a Large-scale Wikipedia-based English Information Retrieval Dataset
Jibril Frej | Didier Schwab | Jean-Pierre Chevallet

Over the past years, deep learning methods allowed for new state-of-the-art results in ad-hoc information retrieval. However such methods usually require large amounts of annotated data to be effective. Since most standard ad-hoc information retrieval datasets publicly available for academic research (e.g. Robust04, ClueWeb09) have at most 250 annotated queries, the recent deep learning models for information retrieval perform poorly on these datasets. These models (e.g. DUET, Conv-KNRM) are trained and evaluated on data collected from commercial search engines not publicly available for academic research which is a problem for reproducibility and the advancement of research. In this paper, we propose WIKIR : an open-source toolkit to automatically build large-scale English information retrieval datasets based on Wikipedia. WIKIR is publicly available on GitHub. We also provide wikIR59k : a large-scale publicly available dataset that contains 59,252 queries and 2,617,003 (query, relevant documents) pairs.

pdf bib
HBCP Corpus : A New Resource for the Analysis of Behavioural Change Intervention ReportsHBCP Corpus: A New Resource for the Analysis of Behavioural Change Intervention Reports
Francesca Bonin | Martin Gleize | Ailbhe Finnerty | Candice Moore | Charles Jochim | Emma Norris | Yufang Hou | Alison J. Wright | Debasis Ganguly | Emily Hayes | Silje Zink | Alessandra Pascale | Pol Mac Aonghusa | Susan Michie

Due to the fast pace at which research reports in behaviour change are published, researchers, consultants and policymakers would benefit from more automatic ways to process these reports. Automatic extraction of the reports’ intervention content, population, settings and their results etc. are essential in synthesising and summarising the literature. However, to the best of our knowledge, no unique resource exists at the moment to facilitate this synthesis. In this paper, we describe the construction of a corpus of published behaviour change intervention evaluation reports aimed at smoking cessation. We also describe and release the annotation of 57 entities, that can be used as an off-the-shelf data resource for tasks such as entity recognition, etc. Both the corpus and the annotation dataset are being made available to the community.

pdf bib
Do not let the history haunt you : Mitigating Compounding Errors in Conversational Question Answering
Angrosh Mandya | James O’ Neill | Danushka Bollegala | Frans Coenen

The Conversational Question Answering (CoQA) task involves answering a sequence of inter-related conversational questions about a contextual paragraph. Although existing approaches employ human-written ground-truth answers for answering conversational questions at test time, in a realistic scenario, the CoQA model will not have any access to ground-truth answers for the previous questions, compelling the model to rely upon its own previously predicted answers for answering the subsequent questions. In this paper, we find that compounding errors occur when using previously predicted answers at test time, significantly lowering the performance of CoQA systems. To solve this problem, we propose a sampling strategy that dynamically selects between target answers and model predictions during training, thereby closely simulating the situation at test time. Further, we analyse the severity of this phenomena as a function of the question type, conversation length and domain type.

pdf bib
CLEEK : A Chinese Long-text Corpus for Entity LinkingCLEEK: A Chinese Long-text Corpus for Entity Linking
Weixin Zeng | Xiang Zhao | Jiuyang Tang | Zhen Tan | Xuqian Huang

Entity linking, as one of the fundamental tasks in natural language processing, is crucial to knowledge fusion, knowledge base construction and update. Nevertheless, in contrast to the research on entity linking for English text, which undergoes continuous development, the Chinese counterpart is still in its infancy. One prominent issue lies in publicly available annotated datasets and evaluation benchmarks, which are lacking and deficient. In specific, existing Chinese corpora for entity linking were mainly constructed from noisy short texts, such as microblogs and news headings, where long texts were largely overlooked, which yet constitute a wider spectrum of real-life scenarios. To address the issue, in this work, we build CLEEK, a Chinese corpus of multi-domain long text for entity linking, in order to encourage advancement of entity linking in languages besides English. The corpus consists of 100 documents from diverse domains, and is publicly accessible. Moreover, we devise a measure to evaluate the difficulty of documents with respect to entity linking, which is then used to characterize the corpus. Additionally, the results of two baselines and seven state-of-the-art solutions on CLEEK are reported and compared. The empirical results validate the usefulness of CLEEK and the effectiveness of proposed difficulty measure.

pdf bib
Event Extraction from Unstructured Amharic TextAmharic Text
Ephrem Tadesse | Rosa Tsegaye | Kuulaa Qaqqabaa

In information extraction, event extraction is one of the types that extract the specific knowledge of certain incidents from texts. Event extraction has been done on different languages text but not on one of the Semitic language, Amharic. In this study, we present a system that extracts an event from unstructured Amharic text. The system has designed by the integration of supervised machine learning and rule-based approaches. We call this system a hybrid system. The system uses the supervised machine learning to detect events from the text and the handcrafted and the rule-based rules to extract the event from the text. For the event extraction, we have been using event arguments. Event arguments identify event triggering words or phrases that clearly express the occurrence of the event. The event argument attributes can be verbs, nouns, sometimes adjectives (such as rg / wedding) and time as well. The hybrid system has compared with the standalone rule-based method that is well known for event extraction. The study has shown that the hybrid system has outperformed the standalone rule-based method.

pdf bib
Towards Entity Spaces
Marieke van Erp | Paul Groth

Entities are a central element of knowledge bases and are important input to many knowledge-centric tasks including text analysis. For example, they allow us to find documents relevant to a specific entity irrespective of the underlying syntactic expression within a document. However, the entities that are commonly represented in knowledge bases are often a simplification of what is truly being referred to in text. For example, in a knowledge base, we may have an entity for Germany as a country but not for the more fuzzy concept of Germany that covers notions of German Population, German Drivers, and the German Government. Inspired by recent advances in contextual word embeddings, we introduce the concept of entity spaces-specific representations of a set of associated entities with near-identity. Thus, these entity spaces provide a handle to an amorphous grouping of entities. We developed a proof-of-concept for English showing how, through the introduction of entity spaces in the form of disambiguation pages, the recall of entity linking can be improved.

pdf bib
Populating Legal Ontologies using Semantic Role Labeling
Llio Humphreys | Guido Boella | Luigi Di Caro | Livio Robaldo | Leon van der Torre | Sepideh Ghanavati | Robert Muthuri

This paper is concerned with the goal of maintaining legal information and compliance systems : the ‘resource consumption bottleneck’ of creating semantic technologies manually. The use of automated information extraction techniques could significantly reduce this bottleneck. The research question of this paper is : How to address the resource bottleneck problem of creating specialist knowledge management systems? In particular, how to semi-automate the extraction of norms and their elements to populate legal ontologies? This paper shows that the acquisition paradox can be addressed by combining state-of-the-art general-purpose NLP modules with pre- and post-processing using rules based on domain knowledge. It describes a Semantic Role Labeling based information extraction system to extract norms from legislation and represent them as structured norms in legal ontologies. The output is intended to help make laws more accessible, understandable, and searchable in legal document management systems such as Eunomos (Boella et al., 2016).

pdf bib
Domain Adapted Distant Supervision for Pedagogically Motivated Relation Extraction
Oscar Sainz | Oier Lopez de Lacalle | Itziar Aldabe | Montse Maritxalar

In this paper we present a relation extraction system that given a text extracts pedagogically motivated relation types, as a previous step to obtaining a semantic representation of the text which will make possible to automatically generate questions for reading comprehension. The system maps pedagogically motivated relations with relations from ConceptNet and deploys Distant Supervision for relation extraction. We run a study on a subset of those relationships in order to analyse the viability of our approach. For that, we build a domain-specific relation extraction system and explore two relation extraction models : a state-of-the-art model based on transfer learning and a discrete feature based machine learning model. Experiments show that the neural model obtains better results in terms of F-score and we yield promising results on the subset of relations suitable for pedagogical purposes. We thus consider that distant supervision for relation extraction is a valid approach in our target domain, i.e. biology.

pdf bib
NLP Analytics in Finance with DoRe : A French 250 M Tokens Corpus of Corporate Annual ReportsNLP Analytics in Finance with DoRe: A French 250M Tokens Corpus of Corporate Annual Reports
Corentin Masson | Patrick Paroubek

Recent advances in neural computing and word embeddings for semantic processing open many new applications areas which had been left unaddressed so far because of inadequate language understanding capacity. But this new kind of approaches rely even more on training data to be operational. Corpora for financial applications exists, but most of them concern stock market prediction and are in English. To address this need for the French language and regulation oriented applications which require a deeper understanding of the text content, we hereby present DoRe, a French and dialectal French Corpus for NLP analytics in Finance, Regulation and Investment. This corpus is composed of : (a) 1769 Annual Reports from 336 companies among the most capitalized companies in : France (Euronext Paris) & Belgium (Euronext Brussels), covering a time frame from 2009 to 2019, and (b) related MetaData containing information for each company about its ISIN code, capitalization and sector. This corpus is designed to be as modular as possible in order to allow for maximum reuse in different tasks pertaining to Economics, Finance and Regulation. After presenting existing resources, we relate the construction of the DoRe corpus and the rationale behind our choices, concluding on the spectrum of possible uses of this new resource for NLP applications.

pdf bib
Evaluation Dataset and Methodology for Extracting Application-Specific Taxonomies from the Wikipedia Knowledge GraphWikipedia Knowledge Graph
Georgeta Bordea | Stefano Faralli | Fleur Mougin | Paul Buitelaar | Gayo Diallo

In this work, we address the task of extracting application-specific taxonomies from the category hierarchy of Wikipedia. Previous work on pruning the Wikipedia knowledge graph relied on silver standard taxonomies which can only be automatically extracted for a small subset of domains rooted in relatively focused nodes, placed at an intermediate level in the knowledge graphs. In this work, we propose an iterative methodology to extract an application-specific gold standard dataset from a knowledge graph and an evaluation framework to comparatively assess the quality of noisy automatically extracted taxonomies. We employ an existing state of the art algorithm in an iterative manner and we propose several sampling strategies to reduce the amount of manual work needed for evaluation. A first gold standard dataset is released to the research community for this task along with a companion evaluation framework. This dataset addresses a real-world application from the medical domain, namely the extraction of food-drug and herb-drug interactions.

pdf bib
Subjective Evaluation of Comprehensibility in Movie Interactions
Estelle Randria | Lionel Fontan | Maxime Le Coz | Isabelle Ferrané | Julien Pinquier

Various research works have dealt with the comprehensibility of textual, audio, or audiovisual documents, and showed that factors related to text (e.g. linguistic complexity), sound (e.g. speech intelligibility), image (e.g. presence of visual context), or even to cognition and emotion can play a major role in the ability of humans to understand the semantic and pragmatic contents of a given document. However, to date, no reference human data is available that could help investigating the role of the linguistic and extralinguistic information present at these different levels (i.e., linguistic, audio / phonetic, and visual) in multimodal documents (e.g., movies). The present work aimed at building a corpus of human annotations that would help to study further how much and in which way the human perception of comprehensibility (i.e., of the difficulty of comprehension, referred in this paper as overall difficulty) of audiovisual documents is affected (1) by lexical complexity, grammatical complexity, and speech intelligibility, and (2) by the modality / ies (text, audio, video) available to the human recipient.

pdf bib
Prtlet : A Hungarian Corpus of Propaganda Texts from the Hungarian Socialist EraPártélet: A Hungarian Corpus of Propaganda Texts from the Hungarian Socialist Era
Zoltán Kmetty | Veronika Vincze | Dorottya Demszky | Orsolya Ring | Balázs Nagy | Martina Katalin Szabó

In this paper, we present Prtlet, a digitized Hungarian corpus of Communist propaganda texts. Prtlet was the official journal of the governing party during the Hungarian socialism from 1956 to 1989, hence it represents the direct political agitation and propaganda of the dictatorial system in question. The paper has a dual purpose : first, to present a general review of the corpus compilation process and the basic statistical data of the corpus, and second, to demonstrate through two case studies what the dataset can be used for. We show that our corpus provides a unique opportunity for conducting research on Hungarian propaganda discourse, as well as analyzing changes of this discourse over a 35-year period of time with computer-assisted methods.

pdf bib
Processing South Asian Languages Written in the Latin Script : the Dakshina DatasetSouth Asian Languages Written in the Latin Script: the Dakshina Dataset
Brian Roark | Lawrence Wolf-Sonkin | Christo Kirov | Sabrina J. Mielke | Cibu Johny | Isin Demirsahin | Keith Hall

This paper describes the Dakshina dataset, a new resource consisting of text in both the Latin and native scripts for 12 South Asian languages. The dataset includes, for each language : 1) native script Wikipedia text ; 2) a romanization lexicon ; and 3) full sentence parallel data in both a native script of the language and the basic Latin alphabet. We document the methods used for preparation and selection of the Wikipedia text in each language ; collection of attested romanizations for sampled lexicons ; and manual romanization of held-out sentences from the native script collections. We additionally provide baseline results on several tasks made possible by the dataset, including single word transliteration, full sentence transliteration, and language modeling of native script and romanized text.

pdf bib
Adaptation of Deep Bidirectional Transformers for Afrikaans LanguageAfrikaans Language
Sello Ralethe

The recent success of pretrained language models in Natural Language Processing has sparked interest in training such models for languages other than English. Currently, training of these models can either be monolingual or multilingual based. In the case of multilingual models, such models are trained on concatenated data of multiple languages. We introduce AfriBERT, a language model for the Afrikaans language based on Bidirectional Encoder Representation from Transformers (BERT). We compare the performance of AfriBERT against multilingual BERT in multiple downstream tasks, namely part-of-speech tagging, named-entity recognition, and dependency parsing. Our results show that AfriBERT improves the current state-of-the-art in most of the tasks we considered, and that transfer learning from multilingual to monolingual model can have a significant performance improvement on downstream tasks. We release the pretrained model for AfriBERT.

pdf bib
FlauBERT : Unsupervised Language Model Pre-training for FrenchFlauBERT: Unsupervised Language Model Pre-training for French
Hang Le | Loïc Vial | Jibril Frej | Vincent Segonne | Maximin Coavoux | Benjamin Lecouteux | Alexandre Allauzen | Benoit Crabbé | Laurent Besacier | Didier Schwab

Language models have become a key step to achieve state-of-the art results in many different Natural Language Processing (NLP) tasks. Leveraging the huge amount of unlabeled texts nowadays available, they provide an efficient way to pre-train continuous word representations that can be fine-tuned for a downstream task, along with their contextualization at the sentence level. This has been widely demonstrated for English using contextualized representations (Dai and Le, 2015 ; Peters et al., 2018 ; Howard and Ruder, 2018 ; Radford et al., 2018 ; Devlin et al., 2019 ; Yang et al., 2019b). In this paper, we introduce and share FlauBERT, a model learned on a very large and heterogeneous French corpus. Models of different sizes are trained using the new CNRS (French National Centre for Scientific Research) Jean Zay supercomputer. We apply our French language models to diverse NLP tasks (text classification, paraphrasing, natural language inference, parsing, word sense disambiguation) and show that most of the time they outperform other pre-training approaches. Different versions of FlauBERT as well as a unified evaluation protocol for the downstream tasks, called FLUE (French Language Understanding Evaluation), are shared to the research community for further reproducible experiments in French NLP.

pdf bib
Modeling Factual Claims with Semantic Frames
Fatma Arslan | Josue Caraballo | Damian Jimenez | Chengkai Li

In this paper, we introduce an extension of the Berkeley FrameNet for the structured and semantic modeling of factual claims. Modeling is a robust tool that can be leveraged in many different tasks such as matching claims to existing fact-checks and translating claims to structured queries. Our work introduces 11 new manually crafted frames along with 9 existing FrameNet frames, all of which have been selected with fact-checking in mind. Along with these frames, we are also providing 2,540 fully annotated sentences, which can be used to understand how these frames are intended to work and to train machine learning models. Finally, we are also releasing our annotation tool to facilitate other researchers to make their own local extensions to FrameNet.

pdf bib
Geographically-Balanced Gigaword Corpora for 50 Language VarietiesGigaword Corpora for 50 Language Varieties
Jonathan Dunn | Ben Adams

While text corpora have been steadily increasing in overall size, even very large corpora are not designed to represent global population demographics. For example, recent work has shown that existing English gigaword corpora over-represent inner-circle varieties from the US and the UK. To correct implicit geographic and demographic biases, this paper uses country-level population demographics to guide the construction of gigaword web corpora. The resulting corpora explicitly match the ground-truth geographic distribution of each language, thus equally representing language users from around the world. This is important because it ensures that speakers of under-resourced language varieties (i.e., Indian English or Algerian French) are represented, both in the corpora themselves but also in derivative resources like word embeddings.

pdf bib
Data Augmentation using Machine Translation for Fake News Detection in the Urdu LanguageUrdu Language
Maaz Amjad | Grigori Sidorov | Alisa Zhila

The task of fake news detection is to distinguish legitimate news articles that describe real facts from those which convey deceiving and fictitious information. As the fake news phenomenon is omnipresent across all languages, it is crucial to be able to efficiently solve this problem for languages other than English. A common approach to this task is supervised classification using features of various complexity. Yet supervised machine learning requires substantial amount of annotated data. For English and a small number of other languages, annotated data availability is much higher, whereas for the vast majority of languages, it is almost scarce. We investigate whether machine translation at its present state could be successfully used as an automated technique for annotated corpora creation and augmentation for fake news detection focusing on the English-Urdu language pair. We train a fake news classifier for Urdu on (1) the manually annotated dataset originally in Urdu and (2) the machine-translated version of an existing annotated fake news dataset originally in English. We show that at the present state of machine translation quality for the English-Urdu language pair, the fully automated data augmentation through machine translation did not provide improvement for fake news detection in Urdu.

pdf bib
Evaluation of Greek Word EmbeddingsGreek Word Embeddings
Stamatis Outsios | Christos Karatsalos | Konstantinos Skianis | Michalis Vazirgiannis

Since word embeddings have been the most popular input for many NLP tasks, evaluating their quality is critical. Most research efforts are focusing on English word embeddings. This paper addresses the problem of training and evaluating such models for the Greek language. We present a new word analogy test set considering the original English Word2vec analogy test set and some specific linguistic aspects of the Greek language as well. Moreover, we create a Greek version of WordSim353 test collection for a basic evaluation of word similarities. Produced resources are available for download. We test seven word vector models and our evaluation shows that we are able to create meaningful representations. Last, we discover that the morphological complexity of the Greek language and polysemy can influence the quality of the resulting word embeddings.

pdf bib
A Dataset of Mycenaean Linear B SequencesMycenaean Linear B Sequences
Katerina Papavassiliou | Gareth Owens | Dimitrios Kosmopoulos

We present our work towards a dataset of Mycenaean Linear B sequences gathered from the Mycenaean inscriptions written in the 13th and 14th century B.C. (c. 1400-1200 B.C.). The dataset contains sequences of Mycenaean words and ideograms according to the rules of the Mycenaean Greek language in the Late Bronze Age. Our ultimate goal is to contribute to the study, reading and understanding of ancient scripts and languages. Focusing on sequences, we seek to exploit the structure of the entire language, not just the Mycenaean vocabulary, to analyse sequential patterns. We use the dataset to experiment on estimating the missing symbols in damaged inscriptions.

pdf bib
Exploring Bilingual Word Embeddings for Hiligaynon, a Low-Resource LanguageHiligaynon, a Low-Resource Language
Leah Michel | Viktor Hangya | Alexander Fraser

This paper investigates the use of bilingual word embeddings for mining Hiligaynon translations of English words. There is very little research on Hiligaynon, an extremely low-resource language of Malayo-Polynesian origin with over 9 million speakers in the Philippines (we found just one paper). We use a publicly available Hiligaynon corpus with only 300 K words, and match it with a comparable corpus in English. As there are no bilingual resources available, we manually develop a English-Hiligaynon lexicon and use this to train bilingual word embeddings. But we fail to mine accurate translations due to the small amount of data. To find out if the same holds true for a related language pair, we simulate the same low-resource setup on English to German and arrive at similar results. We then vary the size of the comparable English and German corpora to determine the minimum corpus size necessary to achieve competitive results. Further, we investigate the role of the seed lexicon. We show that with the same corpus size but with a smaller seed lexicon, performance can surpass results of previous studies. We release the lexicon of 1,200 English-Hiligaynon word pairs we created to encourage further investigation.

pdf bib
Evaluating Sentence Segmentation in Different Datasets of Neuropsychological Language Tests in Brazilian PortugueseBrazilian Portuguese
Edresson Casanova | Marcos Treviso | Lilian Hübner | Sandra Aluísio

Automatic analysis of connected speech by natural language processing techniques is a promising direction for diagnosing cognitive impairments. However, some difficulties still remain : the time required for manual narrative transcription and the decision on how transcripts should be divided into sentences for successful application of parsers used in metrics, such as Idea Density, to analyze the transcripts. The main goal of this paper was to develop a generic segmentation system for narratives of neuropsychological language tests. We explored the performance of our previous single-dataset-trained sentence segmentation architecture in a richer scenario involving three new datasets used to diagnose cognitive impairments, comprising different stories and two types of stimulus presentation for eliciting narratives visual and oral via illustrated story-book and sequence of scenes, and by retelling. Also, we proposed and evaluated three modifications to our previous RCNN architecture : (i) the inclusion of a Linear Chain CRF ; (ii) the inclusion of a self-attention mechanism ; and (iii) the replacement of the LSTM recurrent layer by a Quasi-Recurrent Neural Network layer. Our study allowed us to develop two new models for segmenting impaired speech transcriptions, along with an ideal combination of datasets and specific groups of narratives to be used as the training set.

pdf bib
Development of a Guarani-Spanish Parallel CorpusGuarani - Spanish Parallel Corpus
Luis Chiruzzo | Pedro Amarilla | Adolfo Ríos | Gustavo Giménez Lugo

This paper presents the development of a Guarani-Spanish parallel corpus with sentence-level alignment. The Guarani sentences of the corpus use the Jopara Guarani dialect, the dialect of Guarani spoken in Paraguay, which is based on Guarani grammar and may include several Spanish loanwords or neologisms. The corpus has around 14,500 sentence pairs aligned using a semi-automatic process, containing 228,000 Guarani tokens and 336,000 Spanish tokens extracted from web sources.

pdf bib
Processing Language Resources of Under-Resourced and Endangered Languages for the Generation of Augmentative Alternative Communication Boards
Anne Ferger

Under-resourced and endangered or small languages yield problems for automatic processing and exploiting because of the small amount of available data. This paper shows an approach using different annotations of enriched linguistic research data to create communication boards commonly used in Alternative Augmentative Communication (AAC). Using manually created lexical analysis and rich annotation (instead of high data quantity) allows for an automated creation of AAC communication boards. The example presented in this paper uses data of the indigenous language Dolgan (an endangered Turkic language of Northern Siberia) created in the project INEL(Arkhipov and Dbritz, 2018) to generate a basic communication board with audio snippets to be used in e.g. hospital communication or for multilingual settings. The created boards can be importet into various AAC software. In addition, the usage of standard formats makes this approach applicable to various different use cases.

pdf bib
Towards a Spell Checker for Zamboanga Chavacano OrthographyZamboanga Chavacano Orthography
Marcelo Yuji Himoro | Antonio Pareja-Lora

Zamboanga Chabacano (ZC) is the most vibrant variety of Philippine Creole Spanish, with over 400,000 native speakers in the Philippines (as of 2010). Following its introduction as a subject and a medium of instruction in the public schools of Zamboanga City from Grade 1 to 3 in 2012, an official orthography for this variety-the so-called Zamboanga Chavacano Orthography-has been approved in 2014. Its complexity, however, is a barrier to most speakers, since it does not necessarily reflect the particular phonetic evolution in ZC, but favours etymology instead. The distance between the correct spelling and the different spelling variations is often so great that delivering acceptable performance with the current de facto spell checking technologies may be challenging. The goals of this research have been to propose i) a spelling error taxonomy for ZC, formalised as an ontology and ii) an adaptive spell checking approach using Character-Based Statistical Machine Translation to correct spelling errors in ZC. Our results show that this approach is suitable for the goals mentioned and that it could be combined with other current spell checking technologies to achieve even higher performance.

pdf bib
Automatic Creation of Text Corpora for Low-Resource Languages from the Internet : The Case of Swiss GermanInternet: The Case of Swiss German
Lucy Linder | Michael Jungo | Jean Hennebert | Claudiu Cristian Musat | Andreas Fischer

This paper presents SwissCrawl, the largest Swiss German text corpus to date. Composed of more than half a million sentences, it was generated using a customized web scraping tool that could be applied to other low-resource languages as well. The approach demonstrates how freely available web pages can be used to construct comprehensive text corpora, which are of fundamental importance for natural language processing. In an experimental evaluation, we show that using the new corpus leads to significant improvements for the task of language modeling.

pdf bib
Evaluating the Impact of Sub-word Information and Cross-lingual Word Embeddings on Mi’kmaq Language Modelling
Jeremie Boudreau | Akankshya Patra | Ashima Suvarna | Paul Cook

Mi’kmaq is an Indigenous language spoken primarily in Eastern Canada. It is polysynthetic and low-resource. In this paper we consider a range of n-gram and RNN language models for Mi’kmaq. We find that an RNN language model, initialized with pre-trained fastText embeddings, performs best, highlighting the importance of sub-word information for Mi’kmaq language modelling. We further consider approaches to language modelling that incorporate cross-lingual word embeddings, but do not see improvements with these models. Finally we consider language models that operate over segmentations produced by SentencePiece which include sub-word units as tokens as opposed to word-level models. We see improvements for this approach over word-level language models, again indicating that sub-word modelling is important for Mi’kmaq language modelling.

pdf bib
Massive vs. Curated Embeddings for Low-Resourced Languages : the Case of Yorb and TwiYorùbá and Twi
Jesujoba Alabi | Kwabena Amponsah-Kaakyire | David Adelani | Cristina España-Bonet

The success of several architectures to learn semantic representations from unannotated text and the availability of these kind of texts in online multilingual resources such as Wikipedia has facilitated the massive and automatic creation of resources for multiple languages. The evaluation of such resources is usually done for the high-resourced languages, where one has a smorgasbord of tasks and test sets to evaluate on. For low-resourced languages, the evaluation is more difficult and normally ignored, with the hope that the impressive capability of deep learning architectures to learn (multilingual) representations in the high-resourced setting holds in the low-resourced setting too. In this paper we focus on two African languages, Yorb and Twi, and compare the word embeddings obtained in this way, with word embeddings obtained from curated corpora and a language-dependent processing. We analyse the noise in the publicly available corpora, collect high quality and noisy data for the two languages and quantify the improvements that depend not only on the amount of data but on the quality too. We also use different architectures that learn word representations both from surface forms and characters to further exploit all the available information which showed to be important for these languages. For the evaluation, we manually translate the wordsim-353 word pairs dataset from English into Yorb and Twi. We extend the analysis to contextual word embeddings and evaluate multilingual BERT on a named entity recognition task. For this, we annotate with named entities the Global Voices corpus for Yorb. As output of the work, we provide corpora, embeddings and the test suits for both languages.

pdf bib
TRopBank : Turkish PropBank V2.0TRopBank: Turkish PropBank V2.0
Neslihan Kara | Deniz Baran Aslan | Büşra Marşan | Özge Bakay | Koray Ak | Olcay Taner Yıldız

In this paper, we present and explain TRopBank Turkish PropBank v2.0. PropBank is a hand-annotated corpus of propositions which is used to obtain the predicate-argument information of a language. Predicate-argument information of a language can help understand semantic roles of arguments. Turkish PropBank v2.0, unlike PropBank v1.0, has a much more extensive list of Turkish verbs, with 17.673 verbs in total.

pdf bib
An Empirical Evaluation of Annotation Practices in Corpora from Language Documentation
Kilu von Prince | Sebastian Nordhoff

For most of the world’s languages, no primary data are available, even as many languages are disappearing. Throughout the last two decades, however, language documentation projects have produced substantial amounts of primary data from a wide variety of endangered languages. These resources are still in the early days of their exploration. One of the factors that makes them hard to use is a relative lack of standardized annotation conventions. In this paper, we will describe common practices in existing corpora in order to facilitate their future processing. After a brief introduction of the main formats used for annotation files, we will focus on commonly used tiers in the widespread ELAN and Toolbox formats. Minimally, corpora from language documentation contain a transcription tier and an aligned translation tier, which means they constitute parallel corpora. Additional common annotations include named references, morpheme separation, morpheme-by-morpheme glosses, part-of-speech tags and notes.

pdf bib
Building a Task-oriented Dialog System for Languages with no Training Data : the Case for BasqueBasque
Maddalen López de Lacalle | Xabier Saralegi | Iñaki San Vicente

This paper presents an approach for developing a task-oriented dialog system for less-resourced languages in scenarios where training data is not available. Both intent classification and slot filling are tackled. We project the existing annotations in rich-resource languages by means of Neural Machine Translation (NMT) and posterior word alignments. We then compare training on the projected monolingual data with direct model transfer alternatives. Intent Classifiers and slot filling sequence taggers are implemented using a BiLSTM architecture or by fine-tuning BERT transformer models. Models learnt exclusively from Basque projected data provide better accuracies for slot filling. Combining Basque projected train data with rich-resource languages data outperforms consistently models trained solely on projected data for intent classification. At any rate, we achieve competitive performance in both tasks, with accuracies of 81 % for intent classification and 77 % for slot filling.

pdf bib
A Major Wordnet for a Minority Language : Scottish GaelicWordnet for a Minority Language: Scottish Gaelic
Gábor Bella | Fiona McNeill | Rody Gorman | Caoimhin O Donnaile | Kirsty MacDonald | Yamini Chandrashekar | Abed Alhakim Freihat | Fausto Giunchiglia

We present a new wordnet resource for Scottish Gaelic, a Celtic minority language spoken by about 60,000 speakers, most of whom live in Northwestern Scotland. The wordnet contains over 15 thousand word senses and was constructed by merging ten thousand new, high-quality translations, provided and validated by language experts, with an existing wordnet derived from Wiktionary. This new, considerably extended wordnetcurrently among the 30 largest in the worldtargets multiple communities : language speakers and learners ; linguists ; computer scientists solving problems related to natural language processing. By publishing it as a freely downloadable resource, we hope to contribute to the long-term preservation of Scottish Gaelic as a living language, both offline and on the Web.

pdf bib
Learnings from Technological Interventions in a Low Resource Language : A Case-Study on GondiGondi
Devansh Mehta | Sebastin Santy | Ramaravind Kommiya Mothilal | Brij Mohan Lal Srivastava | Alok Sharma | Anurag Shukla | Vishnu Prasad | Venkanna U | Amit Sharma | Kalika Bali

The primary obstacle to developing technologies for low-resource languages is the lack of usable data. In this paper, we report the adaption and deployment of 4 technology-driven methods of data collection for Gondi, a low-resource vulnerable language spoken by around 2.3 million tribal people in south and central India. In the process of data collection, we also help in its revival by expanding access to information in Gondi through the creation of linguistic resources that can be used by the community, such as a dictionary, children’s stories, an app with Gondi content from multiple sources and an Interactive Voice Response (IVR) based mass awareness platform. At the end of these interventions, we collected a little less than 12,000 translated words and/or sentences and identified more than 650 community members whose help can be solicited for future translation efforts. The larger goal of the project is collecting enough data in Gondi to build and deploy viable language technologies like machine translation and speech to text systems that can help take the language onto the internet.

pdf bib
Optimizing Annotation Effort Using Active Learning Strategies : A Sentiment Analysis Case Study in PersianPersian
Seyed Arad Ashrafi Asli | Behnam Sabeti | Zahra Majdabadi | Preni Golazizian | Reza Fahmi | Omid Momenzadeh

Deep learning models are the current State-of-the-art methodologies towards many real-world problems. However, they need a substantial amount of labeled data to be trained appropriately. Acquiring labeled data can be challenging in some particular domains or less-resourced languages. There are some practical solutions regarding these issues, such as Active Learning and Transfer Learning. Active learning’s idea is simple : let the model choose the samples for annotation instead of labeling the whole dataset. This method leads to a more efficient annotation process. Active Learning models can achieve the baseline performance (the accuracy of the model trained on the whole dataset), with a considerably lower amount of labeled data. Several active learning approaches are tested in this work, and their compatibility with Persian is examined using a brand-new sentiment analysis dataset that is also introduced in this work. MirasOpinion, which to our knowledge is the largest Persian sentiment analysis dataset, is crawled from a Persian e-commerce website and annotated using a crowd-sourcing policy. LDA sampling, which is an efficient Active Learning strategy using Topic Modeling, is proposed in this research. Active Learning Strategies have shown promising results in the Persian language, and LDA sampling showed a competitive performance compared to other approaches.

pdf bib
CA-EHN : Commonsense Analogy from E-HowNetCA-EHN: Commonsense Analogy from E-HowNet
Peng-Hsuan Li | Tsan-Yu Yang | Wei-Yun Ma

Embedding commonsense knowledge is crucial for end-to-end models to generalize inference beyond training corpora. However, existing word analogy datasets have tended to be handcrafted, involving permutations of hundreds of words with only dozens of pre-defined relations, mostly morphological relations and named entities. In this work, we model commonsense knowledge down to word-level analogical reasoning by leveraging E-HowNet, an ontology that annotates 88 K Chinese words with their structured sense definitions and English translations. We present CA-EHN, the first commonsense word analogy dataset containing 90,505 analogies covering 5,656 words and 763 relations. Experiments show that CA-EHN stands out as a great indicator of how well word representations embed commonsense knowledge. The dataset is publicly available at.

pdf bib
Building Semantic Grams of Human Knowledge
Valentina Leone | Giovanni Siragusa | Luigi Di Caro | Roberto Navigli

Word senses are typically defined with textual definitions for human consumption and, in computational lexicons, put in context via lexical-semantic relations such as synonymy, antonymy, hypernymy, etc. In this paper we embrace a radically different paradigm that provides a slot-filler structure, called semagram, to define the meaning of words in terms of their prototypical semantic information. We propose a semagram-based knowledge model composed of 26 semantic relationships which integrates features from a range of different sources, such as computational lexicons and property norms. We describe an annotation exercise regarding 50 concepts over 10 different categories and put forward different automated approaches for extending the semagram base to thousands of concepts. We finally evaluated the impact of the proposed resource on a semantic similarity task, showing significant improvements over state-of-the-art word embeddings.

pdf bib
Introducing Lexical Masks : a New Representation of Lexical Entries for Better Evaluation and Exchange of Lexicons
Bruno Cartoni | Daniel Calvelo Aros | Denny Vrandecic | Saran Lertpradit

The evaluation and exchange of large lexicon databases remains a challenge in many NLP applications. Despite the existence of commonly accepted standards for the format and the features used in a lexicon, there is still a lack of precise and interoperable specification requirements about how lexical entries of a particular language should look like, both in terms of the numbers of forms and in terms of features associated with these forms. This paper presents the notion of lexical masks, a powerful tool used to evaluate and exchange lexicon databases in many languages.

pdf bib
Odi et Amo. Creating, Evaluating and Extending Sentiment Lexicons for Latin.Amo. Creating, Evaluating and Extending Sentiment Lexicons for Latin.
Rachele Sprugnoli | Marco Passarotti | Daniela Corbetta | Andrea Peverelli

Sentiment lexicons are essential for developing automatic sentiment analysis systems, but the resources currently available mostly cover modern languages. Lexicons for ancient languages are few and not evaluated with high-quality gold standards. However, the study of attitudes and emotions in ancient texts is a growing field of research which poses specific issues (e.g., lack of native speakers, limited amount of data, unusual textual genres for the sentiment analysis task, such as philosophical or documentary texts) and can have an impact on the work of scholars coming from several disciplines besides computational linguistics, e.g. historians and philologists. The work presented in this paper aims at providing the research community with a set of sentiment lexicons built by taking advantage of manually-curated resources belonging to the long tradition of Latin corpora and lexicons creation. Our interdisciplinary approach led us to release : i) two automatically generated sentiment lexicons ; ii) a gold standard developed by two Latin language and culture experts ; iii) a silver standard in which semantic and derivational relations are exploited so to extend the list of lexical items of the gold standard. In addition, the evaluation procedure is described together with a first application of the lexicons to a Latin tragedy.

pdf bib
WordWars : A Dataset to Examine the Natural Selection of WordsWordWars: A Dataset to Examine the Natural Selection of Words
Saif M. Mohammad

There is a growing body of work on how word meaning changes over time : mutation. In contrast, there is very little work on how different words compete to represent the same meaning, and how the degree of success of words in that competition changes over time : natural selection. We present a new dataset, WordWars, with historical frequency data from the early 1800s to the early 2000s for monosemous English words in over 5000 synsets. We explore three broad questions with the dataset : (1) what is the degree to which predominant words in these synsets have changed, (2) how do prominent word features such as frequency, length, and concreteness impact natural selection, and (3) what are the differences between the predominant words of the 2000s and the predominant words of early 1800s. We show that close to one third of the synsets undergo a change in the predominant word in this time period. Manual annotation of these pairs shows that about 15 % of these are orthographic variations, 25 % involve affix changes, and 60 % have completely different roots. We find that frequency, length, and concreteness all impact natural selection, albeit in different ways.

pdf bib
Challenge Dataset of Cognates and False Friend Pairs from Indian LanguagesIndian Languages
Diptesh Kanojia | Malhar Kulkarni | Pushpak Bhattacharyya | Gholamreza Haffari

Cognates are present in multiple variants of the same text across different languages (e.g., hund in German and hound in the English language mean dog). They pose a challenge to various Natural Language Processing (NLP) applications such as Machine Translation, Cross-lingual Sense Disambiguation, Computational Phylogenetics, and Information Retrieval. A possible solution to address this challenge is to identify cognates across language pairs. In this paper, we describe the creation of two cognate datasets for twelve Indian languages namely Sanskrit, Hindi, Assamese, Oriya, Kannada, Gujarati, Tamil, Telugu, Punjabi, Bengali, Marathi, and Malayalam. We digitize the cognate data from an Indian language cognate dictionary and utilize linked Indian language Wordnets to generate cognate sets. Additionally, we use the Wordnet data to create a False Friends’ dataset for eleven language pairs. We also evaluate the efficacy of our dataset using previously available baseline cognate detection approaches. We also perform a manual evaluation with the help of lexicographers and release the curated gold-standard dataset with this paper.

pdf bib
Development of a Japanese Personality Dictionary based on Psychological MethodsJapanese Personality Dictionary based on Psychological Methods
Ritsuko Iwai | Daisuke Kawahara | Takatsune Kumada | Sadao Kurohashi

We propose a new approach to constructing a personality dictionary with psychological evidence. In this study, we collect personality words, using word embeddings, and construct a personality dictionary with weights for Big Five traits. The weights are calculated based on the responses of the large sample (N=1,938, female = 1,004, M=49.8years old:20-78, SD=16.3). All the respondents answered a 20-item personality questionnaire and 537 personality items derived from word embeddings. We present the procedures to examine the qualities of responses with psychological methods and to calculate the weights. These result in a personality dictionary with two sub-dictionaries. We also discuss an application of the acquired resources.

pdf bib
A Lexicon-Based Approach for Detecting Hedges in Informal Text
Jumayel Islam | Lu Xiao | Robert E. Mercer

Hedging is a commonly used strategy in conversational management to show the speaker’s lack of commitment to what they communicate, which may signal problems between the speakers. Our project is interested in examining the presence of hedging words and phrases in identifying the tension between an interviewer and interviewee during a survivor interview. While there have been studies on hedging detection in the natural language processing literature, all existing work has focused on structured texts and formal communications. Our project thus investigated a corpus of eight unstructured conversational interviews about the Rwanda Genocide and identified hedging patterns in the interviewees’ responses. Our work produced three manually constructed lists of hedge words, booster words, and hedging phrases. Leveraging these lexicons, we developed a rule-based algorithm that detects sentence-level hedges in informal conversations such as survivor interviews. Our work also produced a dataset of 3000 sentences having the categories Hedge and Non-hedge annotated by three researchers. With experiments on this annotated dataset, we verify the efficacy of our proposed algorithm. Our work contributes to the further development of tools that identify hedges from informal conversations and discussions.

pdf bib
Inducing Universal Semantic Tag Vectors
Da Huo | Gerard de Melo

Given the well-established usefulness of part-of-speech tag annotations in many syntactically oriented downstream NLP tasks, the recently proposed notion of semantic tagging (Bjerva et al. 2016) aims at tagging words with tags informed by semantic distinctions, which are likely to be useful across a range of semantic tasks. To this end, their annotation scheme distinguishes, for instance, privative attributes from subsective ones. While annotated corpora exist, their size is limited and thus many words are out-of-vocabulary words. In this paper, we study to what extent we can automatically predict the tags associated with unseen words. We draw on large-scale word representation data to derive a large new Semantic Tag lexicon. Our experiments show that we can infer semantic tags for words with high accuracy both monolingually and cross-lingually.

pdf bib
Towards a Semi-Automatic Detection of Reflexive and Reciprocal Constructions and Their Representation in a Valency Lexicon
Václava Kettnerová | Marketa Lopatkova | Anna Vernerová | Petra Barancikova

Valency lexicons usually describe valency behavior of verbs in non-reflexive and non-reciprocal constructions. However, reflexive and reciprocal constructions are common morphosyntactic forms of verbs. Both of these constructions are characterized by regular changes in morphosyntactic properties of verbs, thus they can be described by grammatical rules. On the other hand, the possibility to create reflexive and/or reciprocal constructions can not be trivially derived from the morphosyntactic structure of verbs as it is conditioned by their semantic properties as well. A large-coverage valency lexicon allowing for rule based generation of all well formed verb constructions should thus integrate the information on reflexivity and reciprocity. In this paper, we propose a semi-automatic procedure, based on grammatical constraints on reflexivity and reciprocity, detecting those verbs that form reflexive and reciprocal constructions in corpus data. However, exploitation of corpus data for this purpose is complicated due to the diverse functions of reflexive markers crossing the domain of reflexivity and reciprocity. The list of verbs identified by the previous procedure is thus further used in an automatic experiment, applying word embeddings for detecting semantically similar verbs. These candidate verbs have been manually verified and annotation of their reflexive and reciprocal constructions has been integrated into the valency lexicon of Czech verbs VALLEX.

pdf bib
Languages Resources for Poorly Endowed Languages : The Case Study of Classical ArmenianClassical Armenian
Chahan Vidal-Gorène | Aliénor Decours-Perez

Classical Armenian is a poorly endowed language, that despite a great tradition of lexicographical erudition is coping with a lack of resources. Although numerous initiatives exist to preserve the Classical Armenian language, the lack of precise and complete grammatical and lexicographical resources remains. This article offers a situation analysis of the existing resources for Classical Armenian and presents the new digital resources provided on the Calfa platform. The Calfa project gathers existing resources and updates, enriches and enhances their content to offer the richest database for Classical Armenian today. Faced with the challenges specific to a poorly endowed language, the Calfa project is also developing new technologies and solutions to enable preservation, advanced research, and larger systems and developments for the Armenian language

pdf bib
Linking the TUFS Basic Vocabulary to the Open Multilingual WordnetTUFS Basic Vocabulary to the Open Multilingual Wordnet
Francis Bond | Hiroki Nomoto | Luís Morgado da Costa | Arthur Bond

We describe the linking of the TUFS Basic Vocabulary Modules, created for online language learning, with the Open Multilingual Wordnet. The TUFS modules have roughly 500 lexical entries in 30 languages, each with the lemma, a link across the languages, an example sentence, usage notes and sound files. The Open Multilingual Wordnet has 34 languages (11 shared with TUFS) organized into synsets linked by semantic relations, with examples and definitions for some languages. The links can be used to (i) evaluate existing wordnets, (ii) add data to these wordnets and (iii) create new open wordnets for Khmer, Korean, Lao, Mongolian, Russian, Tagalog, Urdua nd Vietnamese

pdf bib
Some Issues with Building a Multilingual WordnetWordnet
Francis Bond | Luis Morgado da Costa | Michael Wayne Goodman | John Philip McCrae | Ahti Lohk

In this paper we discuss the experience of bringing together over 40 different wordnets. We introduce some extensions to the GWA wordnet LMF format proposed in Vossen et al. (2016) and look at how this new information can be displayed. Notable extensions include : confidence, corpus frequency, orthographic variants, lexicalized and non-lexicalized synsets and lemmas, new parts of speech, and more. Many of these extensions already exist in multiple wordnets the challenge was to find a compatible representation. To this end, we introduce a new version of the Open Multilingual Wordnet (Bond and Foster, 2013), that integrates a new set of tools that tests the extensions introduced by this new format, while also ensuring the integrity of the Collaborative Interlingual Index (CILI : Bond et al., 2016), avoiding the same new concept to be introduced through multiple projects.

pdf bib
Automatic Reconstruction of Missing Romanian Cognates and Unattested Latin WordsRomanian Cognates and Unattested Latin Words
Alina Maria Ciobanu | Liviu P. Dinu | Laurentiu Zoicas

Producing related words is a key concern in historical linguistics. Given an input word, the task is to automatically produce either its proto-word, a cognate pair or a modern word derived from it. In this paper, we apply a method for producing related words based on sequence labeling, aiming to fill in the gaps in incomplete cognate sets in Romance languages with Latin etymology (producing Romanian cognates that are missing) and to reconstruct uncertified Latin words. We further investigate an ensemble-based aggregation for combining and re-ranking the word productions of multiple languages.

pdf bib
A Multilingual Evaluation Dataset for Monolingual Word Sense Alignment
Sina Ahmadi | John Philip McCrae | Sanni Nimb | Fahad Khan | Monica Monachini | Bolette Pedersen | Thierry Declerck | Tanja Wissik | Andrea Bellandi | Irene Pisani | Thomas Troelsgård | Sussi Olsen | Simon Krek | Veronika Lipp | Tamás Váradi | László Simon | András Gyorffy | Carole Tiberius | Tanneke Schoonheim | Yifat Ben Moshe | Maya Rudich | Raya Abu Ahmad | Dorielle Lonke | Kira Kovalenko | Margit Langemets | Jelena Kallas | Oksana Dereza | Theodorus Fransen | David Cillessen | David Lindemann | Mikel Alonso | Ana Salgado | José Luis Sancho | Rafael-J. Ureña-Ruiz | Jordi Porta Zamorano | Kiril Simov | Petya Osenova | Zara Kancheva | Ivaylo Radev | Ranka Stanković | Andrej Perdih | Dejan Gabrovsek

Aligning senses across resources and languages is a challenging task with beneficial applications in the field of natural language processing and electronic lexicography. In this paper, we describe our efforts in manually aligning monolingual dictionaries. The alignment is carried out at sense-level for various resources in 15 languages. Moreover, senses are annotated with possible semantic relationships such as broadness, narrowness, relatedness, and equivalence. In comparison to previous datasets for this task, this dataset covers a wide range of languages and resources and focuses on the more challenging task of linking general-purpose language. We believe that our data will pave the way for further advances in alignment and evaluation of word senses by creating new solutions, particularly those notoriously requiring data such as neural networks. Our resources are publicly available at

pdf bib
The ACoLi Dictionary GraphACoLi Dictionary Graph
Christian Chiarcos | Christian Fäth | Maxim Ionov

In this paper, we report the release of the ACoLi Dictionary Graph, a large-scale collection of multilingual open source dictionaries available in two machine-readable formats, a graph representation in RDF, using the OntoLex-Lemon vocabulary, and a simple tabular data format to facilitate their use in NLP tasks, such as translation inference across dictionaries. We describe the mapping and harmonization of the underlying data structures into a unified representation, its serialization in RDF and TSV, and the release of a massive and coherent amount of lexical data under open licenses.

pdf bib
Resources in Underrepresented Languages : Building a Representative Romanian CorpusRomanian Corpus
Ludmila Midrigan - Ciochina | Victoria Boyd | Lucila Sanchez-Ortega | Diana Malancea_Malac | Doina Midrigan | David P. Corina

The effort in the field of Linguistics to develop theories that aim to explain language-dependent effects on language processing is greatly facilitated by the availability of reliable resources representing different languages. This project presents a detailed description of the process of creating a large and representative corpus in Romanian a relatively under-resourced language with unique structural and typological characteristics, that can be used as a reliable language resource for linguistic studies. The decisions that have guided the construction of the corpus, including the type of corpus, its size and component resource files are discussed. Issues related to data collection, data organization and storage, as well as characteristics of the data included in the corpus are described. Currently, the corpus has approximately 5,500,000 tokens originating from written text and 100,000 tokens of spoken language. it includes language samples that represent a wide variety of registers (i.e. written language-16 registers and 5 registers of spoken language), as well as different authors and speakers

pdf bib
The CLARIN Knowledge Centre for Atypical Communication ExpertiseCLARIN Knowledge Centre for Atypical Communication Expertise
Henk van den Heuvel | Nelleke Oostdijk | Caroline Rowland | Paul Trilsbeek

This paper introduces a new CLARIN Knowledge Center which is the K-Centre for Atypical Communication Expertise (ACE for short) which has been established at the Centre for Language and Speech Technology (CLST) at Radboud University. Atypical communication is an umbrella term used here to denote language use by second language learners, people with language disorders or those suffering from language disabilities, but also more broadly by bilinguals and users of sign languages. It involves multiple modalities (text, speech, sign, gesture) and encompasses different developmental stages. ACE closely collaborates with The Language Archive (TLA) at the Max Planck Institute for Psycholinguistics in order to safeguard GDPR-compliant data storage and access. We explain the mission of ACE and show its potential on a number of showcases and a use case.

pdf bib
The European Language Technology Landscape in 2020 : Language-Centric and Human-Centric AI for Cross-Cultural Communication in Multilingual EuropeEuropean Language Technology Landscape in 2020: Language-Centric and Human-Centric AI for Cross-Cultural Communication in Multilingual Europe
Georg Rehm | Katrin Marheinecke | Stefanie Hegele | Stelios Piperidis | Kalina Bontcheva | Jan Hajič | Khalid Choukri | Andrejs Vasiļjevs | Gerhard Backfried | Christoph Prinz | José Manuel Gómez-Pérez | Luc Meertens | Paul Lukowicz | Josef van Genabith | Andrea Lösch | Philipp Slusallek | Morten Irgens | Patrick Gatellier | Joachim Köhler | Laure Le Bars | Dimitra Anastasiou | Albina Auksoriūtė | Núria Bel | António Branco | Gerhard Budin | Walter Daelemans | Koenraad De Smedt | Radovan Garabík | Maria Gavriilidou | Dagmar Gromann | Svetla Koeva | Simon Krek | Cvetana Krstev | Krister Lindén | Bernardo Magnini | Jan Odijk | Maciej Ogrodniczuk | Eiríkur Rögnvaldsson | Mike Rosner | Bolette Pedersen | Inguna Skadiņa | Marko Tadić | Dan Tufiș | Tamás Váradi | Kadri Vider | Andy Way | François Yvon

Multilingualism is a cultural cornerstone of Europe and firmly anchored in the European treaties including full language equality. However, language barriers impacting business, cross-lingual and cross-cultural communication are still omnipresent. Language Technologies (LTs) are a powerful means to break down these barriers. While the last decade has seen various initiatives that created a multitude of approaches and technologies tailored to Europe’s specific needs, there is still an immense level of fragmentation. At the same time, AI has become an increasingly important concept in the European Information and Communication Technology area. For a few years now, AI including many opportunities, synergies but also misconceptions has been overshadowing every other topic. We present an overview of the European LT landscape, describing funding programmes, activities, actions and challenges in the different countries with regard to LT, including the current state of play in industry and the LT market. We present a brief overview of the main LT-related activities on the EU level in the last ten years and develop strategic guidance with regard to four key dimensions.

pdf bib
A Framework for Shared Agreement of Language Tags beyond ISO 639ISO 639
Frances Gillis-Webber | Sabine Tittel

The identification and annotation of languages in an unambiguous and standardized way is essential for the description of linguistic data. It is the prerequisite for machine-based interpretation, aggregation, and re-use of the data with respect to different languages. This makes it a key aspect especially for Linked Data and the multilingual Semantic Web. The standard for language tags is defined by IETF’s BCP 47 and ISO 639 provides the language codes that are the tags’ main constituents. However, for the identification of lesser-known languages, endangered languages, regional varieties or historical stages of a language, the ISO 639 codes are insufficient. Also, the optional language sub-tags compliant with BCP 47 do not offer a possibility fine-grained enough to represent linguistic variation. We propose a versatile pattern that extends the BCP 47 sub-tag ‘privateuse’ and is, thus, able to overcome the limits of BCP 47 and ISO 639. Sufficient coverage of the pattern is demonstrated with the use case of linguistic Linked Data of the endangered Gascon language. We show how to use a URI shortcode for the extended sub-tag, making the length compliant with BCP 47. We achieve this with a web application and API developed to encode and decode the language tag.

pdf bib
A CLARIN Transcription Portal for Interview DataCLARIN Transcription Portal for Interview Data
Christoph Draxler | Henk van den Heuvel | Arjan van Hessen | Silvia Calamai | Louise Corti

In this paper we present a first version of a transcription portal for audio files based on automatic speech recognition (ASR) in various languages. The portal is implemented in the CLARIN resources research network and intended for use by non-technical scholars. We explain the background and interdisciplinary nature of interview data, the perks and quirks of using ASR for transcribing the audio in a research context, the dos and don’ts for optimal use of the portal, and future developments foreseen. The portal is promoted in a range of workshops, but there are a number of challenges that have to be met. These challenges concern privacy issues, ASR quality, and cost, amongst others.

pdf bib
Constructing a Bilingual Hadith Corpus Using a Segmentation Tool
Shatha Altammami | Eric Atwell | Ammar Alsalka

This article describes the process of gathering and constructing a bilingual parallel corpus of Islamic Hadith, which is the set of narratives reporting different aspects of the prophet Muhammad’s life. The corpus data is gathered from the six canonical Hadith collections using a custom segmentation tool that automatically segments and annotates the two Hadith components with 92 % accuracy. This Hadith segmenter minimises the costs of language resource creation and produces consistent results independently from previous knowledge and experiences that usually influence human annotators. The corpus includes more than 10 M tokens and will be freely available via the LREC repository.

pdf bib
Facilitating Corpus Usage : Making Icelandic Corpora More Accessible for Researchers and Language UsersIcelandic Corpora More Accessible for Researchers and Language Users
Steinþór Steingrímsson | Starkaður Barkarson | Gunnar Thor Örnólfsson

We introduce an array of open and accessible tools to facilitate the use of the Icelandic Gigaword Corpus, in the field of Natural Language Processing as well as for students, linguists, sociologists and others benefitting from using large corpora. A KWIC engine, powered by the Swedish Korp tool is adapted to the specifics of the corpus. An n-gram viewer, highly customizable to suit different needs, allows users to study word usage throughout the period of our text collection. A frequency dictionary provides much sought after information about word frequency statistics, computed for each subcorpus as well as aggregate, disambiguating homographs based on their respective lemmas and morphosyntactic tags. Furthermore, we provide n-grams based on the corpus, and a variety of pre-trained word embeddings models, based on word2vec, GloVe, fastText and ELMo. For three of the model types, multiple word embedding models are available trained with different algorithms and using either lemmatised or unlemmatised texts.

pdf bib
Privacy by Design and Language Resources
Pawel Kamocki | Andreas Witt

Privacy by Design (also referred to as Data Protection by Design) is an approach in which solutions and mechanisms addressing privacy and data protection are embedded through the entire project lifecycle, from the early design stage, rather than just added as an additional lawyer to the final product. Formulated in the 1990 by the Privacy Commissionner of Ontario, the principle of Privacy by Design has been discussed by institutions and policymakers on both sides of the Atlantic, and mentioned already in the 1995 EU Data Protection Directive (95/46 / EC). More recently, Privacy by Design was introduced as one of the requirements of the General Data Protection Regulation (GDPR), obliging data controllers to define and adopt, already at the conception phase, appropriate measures and safeguards to implement data protection principles and protect the rights of the data subject. Failing to meet this obligation may result in a hefty fine, as it was the case in the Uniontrad decision by the French Data Protection Authority (CNIL). The ambition of the proposed paper is to analyse the practical meaning of Privacy by Design in the context of Language Resources, and propose measures and safeguards that can be implemented by the community to ensure respect of this principle.

pdf bib
Making Metadata Fit for Next Generation Language Technology Platforms : The Metadata Schema of the European Language GridEuropean Language Grid
Penny Labropoulou | Katerina Gkirtzou | Maria Gavriilidou | Miltos Deligiannis | Dimitris Galanis | Stelios Piperidis | Georg Rehm | Maria Berger | Valérie Mapelli | Michael Rigault | Victoria Arranz | Khalid Choukri | Gerhard Backfried | José Manuel Gómez-Pérez | Andres Garcia-Silva

The current scientific and technological landscape is characterised by the increasing availability of data resources and processing tools and services. In this setting, metadata have emerged as a key factor facilitating management, sharing and usage of such digital assets. In this paper we present ELG-SHARE, a rich metadata schema catering for the description of Language Resources and Technologies (processing and generation services and tools, models, corpora, term lists, etc.), as well as related entities (e.g., organizations, projects, supporting documents, etc.). The schema powers the European Language Grid platform that aims to be the primary hub and marketplace for industry-relevant Language Technology in Europe. ELG-SHARE has been based on various metadata schemas, vocabularies, and ontologies, as well as related recommendations and guidelines.

pdf bib
Semi-supervised Development of ASR Systems for Multilingual Code-switched Speech in Under-resourced LanguagesASR Systems for Multilingual Code-switched Speech in Under-resourced Languages
Astik Biswas | Emre Yilmaz | Febe De Wet | Ewald Van der westhuizen | Thomas Niesler

This paper reports on the semi-supervised development of acoustic and language models for under-resourced, code-switched speech in five South African languages. Two approaches are considered. The first constructs four separate bilingual automatic speech recognisers (ASRs) corresponding to four different language pairs between which speakers switch frequently. The second uses a single, unified, five-lingual ASR system that represents all the languages (English, isiZulu, isiXhosa, Setswana and Sesotho). We evaluate the effectiveness of these two approaches when used to add additional data to our extremely sparse training sets. Results indicate that batch-wise semi-supervised training yields better results than a non-batch-wise approach. Furthermore, while the separate bilingual systems achieved better recognition performance than the unified system, they benefited more from pseudolabels generated by the five-lingual system than from those generated by the bilingual systems.

pdf bib
CLFD : A Novel Vectorization Technique and Its Application in Fake News DetectionCLFD: A Novel Vectorization Technique and Its Application in Fake News Detection
Michail Mersinias | Stergos Afantenos | Georgios Chalkiadakis

In recent years, fake news detection has been an emerging research area. In this paper, we put forward a novel statistical approach for the generation of feature vectors to describe a document. Our so-called class label frequency distance (clfd), is shown experimentally to provide an effective way for boosting the performance of machine learning methods. Specifically, our experiments, carried out in the fake news detection domain, verify that efficient traditional machine learning methods that use our vectorization approach, consistently outperform deep learning methods that use word embeddings for small and medium sized datasets, while the results are comparable for large datasets. In addition, we demonstrate that a novel hybrid method that utilizes both a clfd-boosted logistic regression classifier and a deep learning one, clearly outperforms deep learning methods even in large datasets.

pdf bib
Jamo Pair Encoding : Subcharacter Representation-based Extreme Korean Vocabulary Compression for Efficient Subword TokenizationKorean Vocabulary Compression for Efficient Subword Tokenization
Sangwhan Moon | Naoaki Okazaki

In the context of multilingual language model pre-training, vocabulary size for languages with a broad set of potential characters is an unsolved problem. We propose two algorithms applicable in any unsupervised multilingual pre-training task, increasing the elasticity of budget required for building the vocabulary in Byte-Pair Encoding inspired tokenizers, significantly reducing the cost of supporting Korean in a multilingual model.

pdf bib
Semi-supervised Deep Embedded Clustering with Anomaly Detection for Semantic Frame Induction
Zheng Xin Yong | Tiago Timponi Torrent

Although FrameNet is recognized as one of the most fine-grained lexical databases, its coverage of lexical units is still limited. To tackle this issue, we propose a two-step frame induction process : for a set of lexical units not yet present in Berkeley FrameNet data release 1.7, first remove those that can not fit into any existing semantic frame in FrameNet ; then, assign the remaining lexical units to their correct frames. We also present the Semi-supervised Deep Embedded Clustering with Anomaly Detection (SDEC-AD) modelan algorithm that maps high-dimensional contextualized vector representations of lexical units to a low-dimensional latent space for better frame prediction and uses reconstruction error to identify lexical units that can not evoke frames in FrameNet. SDEC-AD outperforms the state-of-the-art methods in both steps of the frame induction process. Empirical results also show that definitions provide contextual information for representing and characterizing the frame membership of lexical units.

pdf bib
Automated Phonological Transcription of Akkadian Cuneiform TextAkkadian Cuneiform Text
Aleksi Sahala | Miikka Silfverberg | Antti Arppe | Krister Lindén

Akkadian was an East-Semitic language spoken in ancient Mesopotamia. The language is attested on hundreds of thousands of cuneiform clay tablets. Several Akkadian text corpora contain only the transliterated text. In this paper, we investigate automated phonological transcription of the transliterated corpora. The phonological transcription provides a linguistically appealing form to represent Akkadian, because the transcription is normalized according to the grammatical description of a given dialect and explicitly shows the Akkadian renderings for Sumerian logograms. Because cuneiform text does not mark the inflection for logograms, the inflected form needs to be inferred from the sentence context. To the best of our knowledge, this is the first documented attempt to automatically transcribe Akkadian. Using a context-aware neural network model, we are able to automatically transcribe syllabic tokens at near human performance with 96 % recall @ 3, while the logogram transcription remains more challenging at 82 % recall @ 3.

pdf bib
Automatic In-the-wild Dataset Annotation with Deep Generalized Multiple Instance Learning
Joana Correia | Isabel Trancoso | Bhiksha Raj

The automation of the diagnosis and monitoring of speech affecting diseases in real life situations, such as Depression or Parkinson’s disease, depends on the existence of rich and large datasets that resemble real life conditions, such as those collected from in-the-wild multimedia repositories like YouTube. However, the cost of manually labeling these large datasets can be prohibitive. In this work, we propose to overcome this problem by automating the annotation process, without any requirements for human intervention. We formulate the annotation problem as a Multiple Instance Learning (MIL) problem, and propose a novel solution that is based on end-to-end differentiable neural networks. Our solution has the additional advantage of generalizing the MIL framework to more scenarios where the data is stil organized in bags but does not meet the MIL bag label conditions. We demonstrate the performance of the proposed method in labeling the in-the-Wild Speech Medical (WSM) Corpus, using simple textual cues extracted from videos and their metadata. Furthermore we show what is the contribution of each type of textual cues for the final model performance, as well as study the influence of the size of the bags of instances in determining the difficulty of the learning problem

pdf bib
How Much Data Do You Need? About the Creation of a Ground Truth for Black Letter and the Effectiveness of Neural OCROCR
Phillip Benjamin Ströbel | Simon Clematide | Martin Volk

Recent advances in Optical Character Recognition (OCR) and Handwritten Text Recognition (HTR) have led to more accurate textrecognition of historical documents. The Digital Humanities heavily profit from these developments, but they still struggle whenchoosing from the plethora of OCR systems available on the one hand and when defining workflows for their projects on the other hand. In this work, we present our approach to build a ground truth for a historical German-language newspaper published in black letter. Wealso report how we used it to systematically evaluate the performance of different OCR engines. Additionally, we used this ground truthto make an informed estimate as to how much data is necessary to achieve high-quality OCR results. The outcomes of our experimentsshow that HTR architectures can successfully recognise black letter text and that a ground truth size of 50 newspaper pages suffices toachieve good OCR accuracy. Moreover, our models perform equally well on data they have not seen during training, which means thatadditional manual correction for diverging data is superfluous.

pdf bib
On The Performance of Time-Pooling Strategies for End-to-End Spoken Language Identification
Joao Monteiro | Md Jahangir Alam | Tiago Falk

Automatic speech processing applications often have to deal with the problem of aggregating local descriptors (i.e., representations of input speech data corresponding to specific portions across the time dimension) and turning them into a single fixed-dimension representation, known as global descriptor, on top of which downstream classification tasks can be performed. In this paper, we provide an empirical assessment of different time pooling strategies when used with state-of-the-art representation learning models. In particular, insights are provided as to when it is suitable to use simple statistics of local descriptors or when more sophisticated approaches are needed. Here, language identification is used as a case study and a database containing ten oriental languages under varying test conditions (short-duration test recordings, confusing languages, unseen languages) is used. Experiments are performed with classifiers trained on top of global descriptors to provide insights on open-set evaluation performance and show that appropriate selection of such pooling strategies yield embeddings able to outperform well-known benchmark systems as well as previously results based on attention only.

pdf bib
SEDAR : a Large Scale French-English Financial Domain Parallel CorpusSEDAR: a Large Scale French-English Financial Domain Parallel Corpus
Abbas Ghaddar | Phillippe Langlais

This paper describes the acquisition, preprocessing and characteristics of SEDAR, a large scale English-French parallel corpus for the financial domain. Our extensive experiments on machine translation show that SEDAR is essential to obtain good performance on finance. We observe a large gain in the performance of machine translation systems trained on SEDAR when tested on finance, which makes SEDAR suitable to study domain adaptation for neural machine translation. The first release of the corpus comprises 8.6 million high quality sentence pairs that are publicly available for research at

pdf bib
JParaCrawl : A Large Scale Web-Based English-Japanese Parallel CorpusJParaCrawl: A Large Scale Web-Based English-Japanese Parallel Corpus
Makoto Morishita | Jun Suzuki | Masaaki Nagata

Recent machine translation algorithms mainly rely on parallel corpora. However, since the availability of parallel corpora remains limited, only some resource-rich language pairs can benefit from them. We constructed a parallel corpus for English-Japanese, for which the amount of publicly available parallel corpora is still limited. We constructed the parallel corpus by broadly crawling the web and automatically aligning parallel sentences. Our collected corpus, called JParaCrawl, amassed over 8.7 million sentence pairs. We show how it includes a broader range of domains and how a neural machine translation model trained with it works as a good pre-trained model for fine-tuning specific domains. The pre-training and fine-tuning approaches achieved or surpassed performance comparable to model training from the initial state and reduced the training time. Additionally, we trained the model with an in-domain dataset and JParaCrawl to show how we achieved the best performance with them. JParaCrawl and the pre-trained models are freely available online for research purposes.

pdf bib
NMT and PBSMT Error Analyses in English to Brazilian Portuguese Automatic TranslationsNMT and PBSMT Error Analyses in English to Brazilian Portuguese Automatic Translations
Helena Caseli | Marcio Inácio

Machine Translation (MT) is one of the most important natural language processing applications. Independently of the applied MT approach, a MT system automatically generates an equivalent version (in some target language) of an input sentence (in some source language). Recently, a new MT approach has been proposed : neural machine translation (NMT). NMT systems have already outperformed traditional phrase-based statistical machine translation (PBSMT) systems for some pairs of languages. However, any MT approach outputs errors. In this work we present a comparative study of MT errors generated by a NMT system and a PBSMT system trained on the same English Brazilian Portuguese parallel corpus. This is the first study of this kind involving NMT for Brazilian Portuguese. Furthermore, the analyses and conclusions presented here point out the specific problems of NMT outputs in relation to PBSMT ones and also give lots of insights into how to implement automatic post-editing for a NMT system. Finally, the corpora annotated with MT errors generated by both PBSMT and NMT systems are also available.

pdf bib
Better Together : Modern Methods Plus Traditional Thinking in NP AlignmentNP Alignment
Ádám Kovács | Judit Ács | Andras Kornai | Gábor Recski

We study a typical intermediary task to Machine Translation, the alignment of NPs in the bitext. After arguing that the task remains relevant even in an end-to-end paradigm, we present simple, dictionary- and word vector-based baselines and a BERT-based system. Our results make clear that even state of the art systems relying on the best end-to-end methods can be improved by bringing in old-fashioned methods such as stopword removal, lemmatization, and dictionaries

pdf bib
Coursera Corpus Mining and Multistage Fine-Tuning for Improving Lectures TranslationCoursera Corpus Mining and Multistage Fine-Tuning for Improving Lectures Translation
Haiyue Song | Raj Dabre | Atsushi Fujita | Sadao Kurohashi

Lectures translation is a case of spoken language translation and there is a lack of publicly available parallel corpora for this purpose. To address this, we examine a framework for parallel corpus mining which is a quick and effective way to mine a parallel corpus from publicly available lectures at Coursera. Our approach determines sentence alignments, relying on machine translation and cosine similarity over continuous-space sentence representations. We also show how to use the resulting corpora in a multistage fine-tuning based domain adaptation for high-quality lectures translation. For JapaneseEnglish lectures translation, we extracted parallel data of approximately 40,000 lines and created development and test sets through manual filtering for benchmarking translation performance. We demonstrate that the mined corpus greatly enhances the quality of translation when used in conjunction with out-of-domain parallel corpora via multistage training. This paper also suggests some guidelines to gather and clean corpora, mine parallel sentences, address noise in the mined data, and create high-quality evaluation splits. For the sake of reproducibility, we have released our code for parallel data creation.

pdf bib
Being Generous with Sub-Words towards Small NMT ChildrenNMT Children
Arne Defauw | Tom Vanallemeersch | Koen Van Winckel | Sara Szoc | Joachim Van den Bogaert

In the context of under-resourced neural machine translation (NMT), transfer learning from an NMT model trained on a high resource language pair, or from a multilingual NMT (M-NMT) model, has been shown to boost performance to a large extent. In this paper, we focus on so-called cold start transfer learning from an M-NMT model, which means that the parent model is not trained on any of the child data. Such a set-up enables quick adaptation of M-NMT models to new languages. We investigate the effectiveness of cold start transfer learning from a many-to-many M-NMT model to an under-resourced child. We show that sufficiently large sub-word vocabularies should be used for transfer learning to be effective in such a scenario. When adopting relatively large sub-word vocabularies we observe increases in performance thanks to transfer learning from a parent M-NMT model, both when translating to and from the under-resourced language. Our proposed approach involving dynamic vocabularies is both practical and effective. We report results on two under-resourced language pairs, i.e. Icelandic-English and Irish-English.

pdf bib
A Test Set for Discourse Translation from Japanese to EnglishJapanese to English
Masaaki Nagata | Makoto Morishita

We made a test set for Japanese-to-English discourse translation to evaluate the power of context-aware machine translation. For each discourse phenomenon, we systematically collected examples where the translation of the second sentence depends on the first sentence. Compared with a previous study on test sets for English-to-French discourse translation (CITATION), we needed different approaches to make the data because Japanese has zero pronouns and represents different senses in different characters. We improved the translation accuracy using context-aware neural machine translation, and the improvement mainly reflects the betterment of the translation of zero pronouns.

pdf bib
To Case or not to case : Evaluating Casing Methods for Neural Machine Translation
Thierry Etchegoyhen | Harritxu Gete

We present a comparative evaluation of casing methods for Neural Machine Translation, to help establish an optimal pre- and post-processing methodology. We trained and compared system variants on data prepared with the main casing methods available, namely translation of raw data without case normalisation, lowercasing with recasing, truecasing, case factors and inline casing. Machine translation models were prepared on WMT 2017 English-German and English-Turkish datasets, for all translation directions, and the evaluation includes reference metric results as well as a targeted analysis of case preservation accuracy. Inline casing, where case information is marked along lowercased words in the training data, proved to be the optimal approach overall in these experiments.

pdf bib
The MARCELL Legislative CorpusMARCELL Legislative Corpus
Tamás Váradi | Svetla Koeva | Martin Yamalov | Marko Tadić | Bálint Sass | Bartłomiej Nitoń | Maciej Ogrodniczuk | Piotr Pęzik | Verginica Barbu Mititelu | Radu Ion | Elena Irimia | Maria Mitrofan | Vasile Păiș | Dan Tufiș | Radovan Garabík | Simon Krek | Andraz Repar | Matjaž Rihtar | Janez Brank

This article presents the current outcomes of the MARCELL CEF Telecom project aiming to collect and deeply annotate a large comparable corpus of legal documents. The MARCELL corpus includes 7 monolingual sub-corpora (Bulgarian, Croatian, Hungarian, Polish, Romanian, Slovak and Slovenian) containing the total body of respective national legislative documents. These sub-corpora are automatically sentence split, tokenized, lemmatized and morphologically and syntactically annotated. The monolingual sub-corpora are complemented by a thematically related parallel corpus (Croatian-English). The metadata and the annotations are uniformly provided for each language specific sub-corpus. Besides the standard morphosyntactic analysis plus named entity and dependency annotation, the corpus is enriched with the IATE and EUROVOC labels. The file format is CoNLL-U Plus Format, containing the ten columns specific to the CoNLL-U format and four extra columns specific to our corpora. The MARCELL corpora represents a rich and valuable source for further studies and developments in machine learning, cross-lingual terminological data extraction and classification.

pdf bib
Handle with Care : A Case Study in Comparable Corpora Exploitation for Neural Machine Translation
Thierry Etchegoyhen | Harritxu Gete

We present the results of a case study in the exploitation of comparable corpora for Neural Machine Translation. A large comparable corpus for Basque-Spanish was prepared, on the basis of independently-produced news by the Basque public broadcaster EiTB, and we discuss the impact of various techniques to exploit the original data in order to determine optimal variants of the corpus. In particular, we show that filtering in terms of alignment thresholds and length-difference outliers has a significant impact on translation quality. The impact of tags identifying comparable data in the training datasets is also evaluated, with results indicating that this technique might be useful to help the models discriminate noisy information, in the form of informational imbalance between aligned sentences. The final corpus was prepared according to the experimental results and is made available to the scientific community for research purposes.

pdf bib
The FISKM Project : Resources and Tools for Finnish-Swedish Machine Translation and Cross-Linguistic ResearchFISKMÖ Project: Resources and Tools for Finnish-Swedish Machine Translation and Cross-Linguistic Research
Jörg Tiedemann | Tommi Nieminen | Mikko Aulamo | Jenna Kanerva | Akseli Leino | Filip Ginter | Niko Papula

This paper presents FISKM

pdf bib
Multiword Expression aware Neural Machine Translation
Andrea Zaninello | Alexandra Birch

Multiword Expressions (MWEs) are a frequently occurring phenomenon found in all natural languages that is of great importance to linguistic theory, natural language processing applications, and machine translation systems. Neural Machine Translation (NMT) architectures do not handle these expressions well and previous studies have rarely addressed MWEs in this framework. In this work, we show that annotation and data augmentation, using external linguistic resources, can improve both translation of MWEs that occur in the source, and the generation of MWEs on the target, and increase performance by up to 5.09 BLEU points on MWE test sets. We also devise a MWE score to specifically assess the quality of MWE translation which agrees with human evaluation. We make available the MWE score implementation along with MWE-annotated training sets and corpus-based lists of MWEs for reproduction and extension.

pdf bib
Finite State Machine Pattern-Root Arabic Morphological Generator, Analyzer and DiacritizerArabic Morphological Generator, Analyzer and Diacritizer
Maha Alkhairy | Afshan Jafri | David Smith

We describe and evaluate the Finite-State Arabic Morphologizer (FSAM) a concatenative (prefix-stem-suffix) and templatic (root- pattern) morphologizer that generates and analyzes undiacritized Modern Standard Arabic (MSA) words, and diacritizes them. Our bidirectional unified-architecture finite state machine (FSM) is based on morphotactic MSA grammatical rules. The FSM models the root-pattern structure related to semantics and syntax, making it readily scalable unlike stem-tabulations in prevailing systems. We evaluate the coverage and accuracy of our model, with coverage being percentage of words in Tashkeela (a large corpus) that can be analyzed. Accuracy is computed against a gold standard, comprising words and properties, created from the intersection of UD PADT treebank and Tashkeela. Coverage of analysis (extraction of root and properties from word) is 82 %. Accuracy results are : root computed from a word (92 %), word generation from a root (100 %), non-root properties of a word (97 %), and diacritization (84 %). FSAM’s non-root results match or surpass MADAMIRA’s, and root result comparisons are not made because of the concatenative nature of publicly available morphologizers.

pdf bib
An Unsupervised Method for Weighting Finite-state Morphological Analyzers
Amr Keleg | Francis Tyers | Nick Howell | Tommi Pirinen

Morphological analysis is one of the tasks that have been studied for years. Different techniques have been used to develop models for performing morphological analysis. Models based on finite state transducers have proved to be more suitable for languages with low available resources. In this paper, we have developed a method for weighting a morphological analyzer built using finite state transducers in order to disambiguate its results. The method is based on a word2vec model that is trained in a completely unsupervised way using raw untagged corpora and is able to capture the semantic meaning of the words. Most of the methods used for disambiguating the results of a morphological analyzer relied on having tagged corpora that need to manually built. Additionally, the method developed uses information about the token irrespective of its context unlike most of the other techniques that heavily rely on the word’s context to disambiguate its set of candidate analyses.

pdf bib
Glawinette : a Linguistically Motivated Derivational Description of French Acquired from GLAWIGlawinette: a Linguistically Motivated Derivational Description of French Acquired from GLAWI
Nabil Hathout | Franck Sajous | Basilio Calderone | Fiammetta Namer

Glawinette is a derivational lexicon of French that will be used to feed the Dmonette database. It has been created from the GLAWI machine readable dictionary. We collected couples of words from the definitions and the morphological sections of the dictionary and then selected the ones that form regular formal analogies and that instantiate frequent enough formal patterns. The graph structure of the morphological families has then been used to identify for each couple of lexemes derivational patterns that are close to the intuition of the morphologists.

pdf bib
BabyFST-Towards a Finite-State Based Computational Model of Ancient BabylonianBabyFST - Towards a Finite-State Based Computational Model of Ancient Babylonian
Aleksi Sahala | Miikka Silfverberg | Antti Arppe | Krister Lindén

Akkadian is a fairly well resourced extinct language that does not yet have a comprehensive morphological analyzer available. In this paper we describe a general finite-state based morphological model for Babylonian, a southern dialect of the Akkadian language, that can achieve a coverage up to 97.3 % and recall up to 93.7 % on lemmatization and POS-tagging task on token level from a transcribed input. Since Akkadian word forms exhibit a high degree of morphological ambiguity, in that only 20.1 % of running word tokens receive a single unambiguous analysis, we attempt a first pass at weighting our finite-state transducer, using existing extensive Akkadian corpora which have been partially validated for their lemmas and parts-of-speech but not the entire morphological analyses. The resultant weighted finite-state transducer yields a moderate improvement so that for 57.4 % of the word tokens the highest ranked analysis is the correct one. We conclude with a short discussion on how morphological ambiguity in the analysis of Akkadian could be further reduced with improvements in the training data used in weighting the finite-state transducer as well as through other, context-based techniques.

pdf bib
Morphological Analysis and Disambiguation for Gulf Arabic : The Interplay between Resources and MethodsGulf Arabic: The Interplay between Resources and Methods
Salam Khalifa | Nasser Zalmout | Nizar Habash

In this paper we present the first full morphological analysis and disambiguation system for Gulf Arabic. We use an existing state-of-the-art morphological disambiguation system to investigate the effects of different data sizes and different combinations of morphological analyzers for Modern Standard Arabic, Egyptian Arabic, and Gulf Arabic. We find that in very low settings, morphological analyzers help boost the performance of the full morphological disambiguation task. However, as the size of resources increase, the value of the morphological analyzers decreases.

pdf bib
Building the Spanish-Croatian Parallel CorpusSpanish-Croatian Parallel Corpus
Bojana Mikelenić | Marko Tadić

This paper describes the building of the first Spanish-Croatian unidirectional parallel corpus, which has been constructed at the Faculty of Humanities and Social Sciences of the University of Zagreb. The corpus is comprised of eleven Spanish novels and their translations to Croatian done by six different professional translators. All the texts were published between 1999 and 2012. The corpus has more than 2 Mw, with approximately 1 Mw for each language. It was automatically sentence segmented and aligned, as well as manually post-corrected, and contains 71,778 translation units. In order to protect the copyright and to make the corpus available under permissive CC-BY licence, the aligned translation units are shuffled. This limits the usability of the corpus for research of language units at sentence and lower language levels only. There are two versions of the corpus in TMX format that will be available for download through META-SHARE and CLARIN ERIC infrastructure. The former contains plain TMX, while the latter is lemmatised and POS-tagged and stored in the aTMX format.

pdf bib
Morfessor EM+Prune : Improved Subword Segmentation with Expectation Maximization and PruningMorfessor EM+Prune: Improved Subword Segmentation with Expectation Maximization and Pruning
Stig-Arne Grönroos | Sami Virpioja | Mikko Kurimo

Data-driven segmentation of words into subword units has been used in various natural language processing applications such as automatic speech recognition and statistical machine translation for almost 20 years. Recently it has became more widely adopted, as models based on deep neural networks often benefit from subword units even for morphologically simpler languages. In this paper, we discuss and compare training algorithms for a unigram subword model, based on the Expectation Maximization algorithm and lexicon pruning. Using English, Finnish, North Sami, and Turkish data sets, we show that this approach is able to find better solutions to the optimization problem defined by the Morfessor Baseline model than its original recursive training algorithm. The improved optimization also leads to higher morphological segmentation accuracy when compared to a linguistic gold standard. We publish implementations of the new algorithms in the widely-used Morfessor software package.

pdf bib
Machine Learning and Deep Neural Network-Based Lemmatization and Morphosyntactic Tagging for SerbianSerbian
Ranka Stankovic | Branislava Šandrih | Cvetana Krstev | Miloš Utvić | Mihailo Skoric

The training of new tagger models for Serbian is primarily motivated by the enhancement of the existing tagset with the grammatical category of a gender. The harmonization of resources that were manually annotated within different projects over a long period of time was an important task, enabled by the development of tools that support partial automation. The supporting tools take into account different taggers and tagsets. This paper focuses on TreeTagger and spaCy taggers, and the annotation schema alignment between Serbian morphological dictionaries, MULTEXT-East and Universal Part-of-Speech tagset. The trained models will be used to publish the new version of the Corpus of Contemporary Serbian as well as the Serbian literary corpus. The performance of developed taggers were compared and the impact of training set size was investigated, which resulted in around 98 % PoS-tagging precision per token for both new models. The sr_basic annotated dataset will also be published.

pdf bib
Fine-grained Morphosyntactic Analysis and Generation Tools for More Than One Thousand Languages
Garrett Nicolai | Dylan Lewis | Arya D. McCarthy | Aaron Mueller | Winston Wu | David Yarowsky

Exploiting the broad translation of the Bible into the world’s languages, we train and distribute morphosyntactic tools for approximately one thousand languages, vastly outstripping previous distributions of tools devoted to the processing of inflectional morphology. Evaluation of the tools on a subset of available inflectional dictionaries demonstrates strong initial models, supplemented and improved through ensembling and dictionary-based reranking. Likewise, a novel type-to-token based evaluation metric allows us to confirm that models generalize well across rare and common forms alike

pdf bib
CCNet : Extracting High Quality Monolingual Datasets from Web Crawl DataCCNet: Extracting High Quality Monolingual Datasets from Web Crawl Data
Guillaume Wenzek | Marie-Anne Lachaux | Alexis Conneau | Vishrav Chaudhary | Francisco Guzmán | Armand Joulin | Edouard Grave

Pre-training text representations have led to significant improvements in many areas of natural language processing. The quality of these models benefits greatly from the size of the pretraining corpora as long as its quality is preserved. In this paper, we describe an automatic pipeline to extract massive high-quality monolingual datasets from Common Crawl for a variety of languages. Our pipeline follows the data processing introduced in fastText (Mikolov et al., 2017 ; Grave et al., 2018), that deduplicates documents and identifies their language. We augment this pipeline with a filtering step to select documents that are close to high quality corpora like Wikipedia.

pdf bib
Multilingual Culture-Independent Word Analogy Datasets
Matej Ulčar | Kristiina Vaik | Jessica Lindström | Milda Dailidėnaitė | Marko Robnik-Šikonja

In text processing, deep neural networks mostly use word embeddings as an input. Embeddings have to ensure that relations between words are reflected through distances in a high-dimensional numeric space. To compare the quality of different text embeddings, typically, we use benchmark datasets. We present a collection of such datasets for the word analogy task in nine languages : Croatian, English, Estonian, Finnish, Latvian, Lithuanian, Russian, Slovenian, and Swedish. We designed the monolingual analogy task to be much more culturally independent and also constructed cross-lingual analogy datasets for the involved languages. We present basic statistics of the created datasets and their initial evaluation using fastText embeddings.

pdf bib
UniSent : Universal Adaptable Sentiment Lexica for 1000 + LanguagesUniSent: Universal Adaptable Sentiment Lexica for 1000+ Languages
Ehsaneddin Asgari | Fabienne Braune | Benjamin Roth | Christoph Ringlstetter | Mohammad Mofrad

In this paper, we introduce UniSent universal sentiment lexica for 1000 + languages. Sentiment lexica are vital for sentiment analysis in absence of document-level annotations, a very common scenario for low-resource languages. To the best of our knowledge, UniSent is the largest sentiment resource to date in terms of the number of covered languages, including many low resource ones. In this work, we use a massively parallel Bible corpus to project sentiment information from English to other languages for sentiment analysis on Twitter data. We introduce a method called DomDrift to mitigate the huge domain mismatch between Bible and Twitter by a confidence weighting scheme that uses domain-specific embeddings to compare the nearest neighbors for a candidate sentiment word in the source (Bible) and target (Twitter) domain. We evaluate the quality of UniSent in a subset of languages for which manually created ground truth was available, Macedonian, Czech, German, Spanish, and French. We show that the quality of UniSent is comparable to manually created sentiment resources when it is used as the sentiment seed for the task of word sentiment prediction on top of embedding representations. In addition, we show that emoticon sentiments could be reliably predicted in the Twitter domain using only UniSent and monolingual embeddings in German, Spanish, French, and Italian. With the publication of this paper, we release the UniSent sentiment lexica at

pdf bib
A Dataset for Multi-lingual Epidemiological Event Extraction
Stephen Mutuvi | Antoine Doucet | Gaël Lejeune | Moses Odeo

This paper proposes a corpus for the development and evaluation of tools and techniques for identifying emerging infectious disease threats in online news text. The corpus can not only be used for information extraction, but also for other natural language processing (NLP) tasks such as text classification. We make use of articles published on the Program for Monitoring Emerging Diseases (ProMED) platform, which provides current information about outbreaks of infectious diseases globally. Among the key pieces of information present in the articles is the uniform resource locator (URL) to the online news sources where the outbreaks were originally reported. We detail the procedure followed to build the dataset, which includes leveraging the source URLs to retrieve the news reports and subsequently pre-processing the retrieved documents. We also report on experimental results of event extraction on the dataset using the Data Analysis for Information Extraction in any Language(DAnIEL) system. DAnIEL is a multilingual news surveillance system that leverages unique attributes associated with news reporting to extract events : repetition and saliency. The system has wide geographical and language coverage, including low-resource languages. In addition, we compare different classification approaches in terms of their ability to differentiate between epidemic-related and unrelated news articles that constitute the corpus.

pdf bib
Analysis of GlobalPhone and Ethiopian Languages Speech Corpora for Multilingual ASRGlobalPhone and Ethiopian Languages Speech Corpora for Multilingual ASR
Martha Yifiru Tachbelie | Solomon Teferra Abate | Tanja Schultz

In this paper, we present the analysis of GlobalPhone (GP) and speech corpora of Ethiopian languages (Amharic, Tigrigna, Oromo and Wolaytta). The aim of the analysis is to select speech data from GP for the development of multilingual Automatic Speech Recognition (ASR) system for the Ethiopian languages. To this end, phonetic overlaps among GP and Ethiopian languages have been analyzed. The result of our analysis shows that there is much phonetic overlap among Ethiopian languages although they are from three different language families. From GP, Turkish, Uyghur and Croatian are found to have much overlap with the Ethiopian languages. On the other hand, Korean has less phonetic overlap with the rest of the languages. Moreover, morphological complexity of the GP and Ethiopian languages, reflected by type to token ration (TTR) and out of vocabulary (OOV) rate, has been analyzed. Both metrics indicated the morphological complexity of the languages. Korean and Amharic have been identified as extremely morphologically complex compared to the other languages. Tigrigna, Russian, Turkish, Polish, etc. are also among the morphologically complex languages.

pdf bib
Multilingualization of Medical Terminology : Semantic and Structural Embedding Approaches
Long-Huei Chen | Kyo Kageura

The multilingualization of terminology is an essential step in the translation pipeline, to ensure the correct transfer of domain-specific concepts. Many institutions and language service providers construct and maintain multilingual terminologies, which constitute important assets. However, the curation of such multilingual resources requires significant human effort ; though automatic multilingual term extraction methods have been proposed so far, they are of limited success as term translation can not be satisfied by simply conveying meaning, but requires the terminologists and domain experts’ knowledge to fit the term within the existing terminology. Here we propose a method to encode the structural property of a term by aligning their embeddings using graph convolutional networks trained from separate languages. We observe that the structural information can augment the semantic methods also explored in this work, and recognize the unique nature of terminologies allows our method to fully take advantage and produce superior results.

pdf bib
CoVoST : A Diverse Multilingual Speech-To-Text Translation CorpusCoVoST: A Diverse Multilingual Speech-To-Text Translation Corpus
Changhan Wang | Juan Pino | Anne Wu | Jiatao Gu

Spoken language translation has recently witnessed a resurgence in popularity, thanks to the development of end-to-end models and the creation of new corpora, such as Augmented LibriSpeech and MuST-C. Existing datasets involve language pairs with English as a source language, involve very specific domains or are low resource. We introduce CoVoST, a multilingual speech-to-text translation corpus from 11 languages into English, diversified with over 11,000 speakers and over 60 accents. We describe the dataset creation methodology and provide empirical evidence of the quality of the data. We also provide initial benchmarks, including, to our knowledge, the first end-to-end many-to-one multilingual models for spoken language translation. CoVoST is released under CC0 license and free to use. We also provide additional evaluation data derived from Tatoeba under CC licenses.

pdf bib
Massively Multilingual Pronunciation Modeling with WikiPronWikiPron
Jackson L. Lee | Lucas F.E. Ashby | M. Elizabeth Garza | Yeonju Lee-Sikka | Sean Miller | Alan Wong | Arya D. McCarthy | Kyle Gorman

We introduce WikiPron, an open-source command-line tool for extracting pronunciation data from Wiktionary, a collaborative multilingual online dictionary. We first describe the design and use of WikiPron. We then discuss the challenges faced scaling this tool to create an automatically-generated database of 1.7 million pronunciations from 165 languages. Finally, we validate the pronunciation database by using it to train and evaluating a collection of generic grapheme-to-phoneme models. The software, pronunciation data, and models are all made available under permissive open-source licenses.

pdf bib
HELFI : a Hebrew-Greek-Finnish Parallel Bible Corpus with Cross-Lingual Morpheme AlignmentHELFI: a Hebrew-Greek-Finnish Parallel Bible Corpus with Cross-Lingual Morpheme Alignment
Anssi Yli-Jyrä | Josi Purhonen | Matti Liljeqvist | Arto Antturi | Pekka Nieminen | Kari M. Räntilä | Valtter Luoto

Twenty-five years ago, morphologically aligned Hebrew-Finnish and Greek-Finnish bitexts (texts accompanied by a translation) were constructed manually in order to create an analytical concordance (Luoto et al., eds. 1997) for a Finnish Bible translation. The creators of the bitexts recently secured the publisher’s permission to release its fine-grained alignment, but the alignment was still dependent on proprietary, third-party resources such as a copyrighted text edition and proprietary morphological analyses of the source texts. In this paper, we describe a nontrivial editorial process starting from the creation of the original one-purpose database and ending with its reconstruction using only freely available text editions and annotations. This process produced an openly available dataset that contains (i) the source texts and their translations, (ii) the morphological analyses, (iii) the cross-lingual morpheme alignments.

pdf bib
Visual Grounding Annotation of Recipe Flow Graph
Taichi Nishimura | Suzushi Tomori | Hayato Hashimoto | Atsushi Hashimoto | Yoko Yamakata | Jun Harashima | Yoshitaka Ushiku | Shinsuke Mori

In this paper, we provide a dataset that gives visual grounding annotations to recipe flow graphs. A recipe flow graph is a representation of the cooking workflow, which is designed with the aim of understanding the workflow from natural language processing. Such a workflow will increase its value when grounded to real-world activities, and visual grounding is a way to do so. Visual grounding is provided as bounding boxes to image sequences of recipes, and each bounding box is linked to an element of the workflow. Because the workflows are also linked to the text, this annotation gives visual grounding with workflow’s contextual information between procedural text and visual observation in an indirect manner. We subsidiarily annotated two types of event attributes with each bounding box : doing-the-action, or done-the-action. As a result of the annotation, we got 2,300 bounding boxes in 272 flow graph recipes. Various experiments showed that the proposed dataset enables us to estimate contextual information described in recipe flow graphs from an image sequence.

pdf bib
Offensive Video Detection : Dataset and Baseline Results
Cleber Alcântara | Viviane Moreira | Diego Feijo

Web-users produce and publish high volumes of data of various types, such as text, images, and videos. The platforms try to restrain their users from publishing offensive content to keep a friendly and respectful environment and rely on moderators to filter the posts. However, this method is insufficient due to the high volume of publications. The identification of offensive material can be performed automatically using machine learning, which needs annotated datasets. Among the published datasets in this matter, the Portuguese language is underrepresented, and videos are little explored. We investigated the problem of offensive video detection by assembling and publishing a dataset of videos in Portuguese containing mostly textual features. We ran experiments using popular machine learning classifiers used in this domain and reported our findings, alongside multiple evaluation metrics. We found that using word embedding with Deep Learning classifiers achieved the best results on average. CNN architectures, Naive Bayes, and Random Forest ranked top among different experiments. Transfer Learning models outperformed Classic algorithms when processing video transcriptions, but scored lower using other feature sets. These findings can be used as a baseline for future works on this subject.

pdf bib
A Domain-Specific Dataset of Difficulty Ratings for German Noun Compounds in the Domains DIY, Cooking and AutomotiveGerman Noun Compounds in the Domains DIY, Cooking and Automotive
Julia Bettinger | Anna Hätty | Michael Dorna | Sabine Schulte im Walde

We present a dataset with difficulty ratings for 1,030 German closed noun compounds extracted from domain-specific texts for do-it-ourself (DIY), cooking and automotive. The dataset includes two-part compounds for cooking and DIY, and two- to four-part compounds for automotive. The compounds were identified in text using the Simple Compound Splitter (Weller-Di Marco, 2017) ; a subset was filtered and balanced for frequency and productivity criteria as basis for manual annotation and fine-grained interpretation. This study presents the creation, the final dataset with ratings from 20 annotators and statistics over the dataset, to provide insight into the perception of domain-specific term difficulty. It is particularly striking that annotators agree on a coarse, binary distinction between easy vs. difficult domain-specific compounds but that a more fine grained distinction of difficulty is not meaningful. We finally discuss the challenges of an annotation for difficulty, which includes both the task description as well as the selection of the data basis.

pdf bib
All That Glitters is Not Gold : A Gold Standard of Adjective-Noun Collocations for GermanGerman
Yana Strakatova | Neele Falk | Isabel Fuhrmann | Erhard Hinrichs | Daniela Rossmann

In this paper we present the GerCo dataset of adjective-noun collocations for German, such as alter Freund ‘old friend’ and tiefe Liebe ‘deep love’. The annotation has been performed by experts based on the annotation scheme introduced in this paper. The resulting dataset contains 4,732 positive and negative instances of collocations and covers all the 16 semantic classes of adjectives as defined in the German wordnet GermaNet. The dataset can serve as a reliable empirical basis for comparing different theoretical frameworks concerned with collocations or as material for data-driven approaches to the studies of collocations including different machine learning experiments. This paper addresses the latter issue by using the GerCo dataset for evaluating different models on the task of automatic collocation identification. We compare lexical association measures with static and contextualized word embeddings. The experiments show that word embeddings outperform methods based on statistical association measures by a wide margin.

pdf bib
A Joint Approach to Compound Splitting and Idiomatic Compound Detection
Irina Krotova | Sergey Aksenov | Ekaterina Artemova

Applications such as machine translation, speech recognition, and information retrieval require efficient handling of noun compounds as they are one of the possible sources for out of vocabulary words. In-depth processing of noun compounds requires not only splitting them into smaller components (or even roots) but also the identification of instances that should remain unsplitted as they are of idiomatic nature. We develop a two-fold deep learning-based approach of noun compound splitting and idiomatic compound detection for the German language that we train using a newly collected corpus of annotated German compounds. Our neural noun compound splitter operates on a sub-word level and outperforms the current state of the art by about 5 %

pdf bib
A Dataset of German Legal Documents for Named Entity RecognitionGerman Legal Documents for Named Entity Recognition
Elena Leitner | Georg Rehm | Julian Moreno-Schneider

We describe a dataset developed for Named Entity Recognition in German federal court decisions. It consists of approx. 67,000 sentences with over 2 million tokens. The resource contains 54,000 manually annotated entities, mapped to 19 fine-grained semantic classes : person, judge, lawyer, country, city, street, landscape, organization, company, institution, court, brand, law, ordinance, European legal norm, regulation, contract, court decision, and legal literature. The legal documents were, furthermore, automatically annotated with more than 35,000 TimeML-based time expressions. The dataset, which is available under a CC-BY 4.0 license in the CoNNL-2002 format, was developed for training an NER service for German legal documents in the EU project Lynx.

pdf bib
Named Entities in Medical Case Reports : Corpus and Experiments
Sarah Schulz | Jurica Ševa | Samuel Rodriguez | Malte Ostendorff | Georg Rehm

We present a new corpus comprising annotations of medical entities in case reports, originating from PubMed Central’s open access library. In the case reports, we annotate cases, conditions, findings, factors and negation modifiers. Moreover, where applicable, we annotate relations between these entities. As such, this is the first corpus of this kind made available to the scientific community in English. It enables the initial investigation of automatic information extraction from case reports through tasks like Named Entity Recognition, Relation Extraction and (sentence / paragraph) relevance detection. Additionally, we present four strong baseline systems for the detection of medical entities made available through the annotated dataset.

pdf bib
Building OCR / NER Test CollectionsOCR/NER Test Collections
Dawn Lawrie | James Mayfield | David Etter

Named entity recognition (NER) identifies spans of text that contain names. Many researchers have reported the results of NER on text created through optical character recognition (OCR) over the past two decades. Unfortunately, the test collections that support this research are annotated with named entities after optical character recognition (OCR) has been run. This means that the collection must be re-annotated if the OCR output changes. Instead by tying annotations to character locations on the page, a collection can be built that supports OCR and NER research without requiring re-annotation when either improves. This means that named entities are annotated on the transcribed text. The transcribed text is all that is needed to evaluate the performance of OCR. For NER evaluation, the tagged OCR output is aligned to the transcriptions the aligned files, creating modified files of each, which are scored. This paper presents a methodology for building such a test collection and releases a collection of Chinese OCR-NER data constructed using the methodology. The paper provides performance baselines for current OCR and NER systems applied to this new collection.

pdf bib
Video Caption Dataset for Describing Human Actions in JapaneseJapanese
Yutaro Shigeto | Yuya Yoshikawa | Jiaqing Lin | Akikazu Takeuchi

In recent years, automatic video caption generation has attracted considerable attention. This paper focuses on the generation of Japanese captions for describing human actions. While most currently available video caption datasets have been constructed for English, there is no equivalent Japanese dataset. To address this, we constructed a large-scale Japanese video caption dataset consisting of 79,822 videos and 399,233 captions. Each caption in our dataset describes a video in the form of who does what and where. To describe human actions, it is important to identify the details of a person, place, and action. Indeed, when we describe human actions, we usually mention the scene, person, and action. In our experiments, we evaluated two caption generation methods to obtain benchmark results. Further, we investigated whether those generation methods could specify who does what and where.

pdf bib
Best Student Forcing : A Simple Training Mechanism in Adversarial Language Generation
Jonathan Sauder | Ting Hu | Xiaoyin Che | Goncalo Mordido | Haojin Yang | Christoph Meinel

Language models trained with Maximum Likelihood Estimation (MLE) have been considered as a mainstream solution in Natural Language Generation (NLG) for years. Recently, various approaches with Generative Adversarial Nets (GANs) have also been proposed. While offering exciting new prospects, GANs in NLG by far are nevertheless reportedly suffering from training instability and mode collapse, and therefore outperformed by conventional MLE models. In this work, we propose techniques for improving GANs in NLG, namely Best Student Forcing (BSF), a novel yet simple adversarial training mechanism in which generated sequences of high quality are selected as temporary ground-truth to further train the generator. We also use an ensemble of discriminators to increase training stability and sample diversity. Evaluation shows that the combination of BSF and multiple discriminators consistently performs better than previous GAN approaches over various metrics, and outperforms a baseline MLE in terms of Fr ech et Distance, a recently proposed metric capturing both sample quality and diversity.

pdf bib
Exploring Transformer Text Generation for Medical Dataset Augmentation
Ali Amin-Nejad | Julia Ive | Sumithra Velupillai

Natural Language Processing (NLP) can help unlock the vast troves of unstructured data in clinical text and thus improve healthcare research. However, a big barrier to developments in this field is data access due to patient confidentiality which prohibits the sharing of this data, resulting in small, fragmented and sequestered openly available datasets. Since NLP model development requires large quantities of data, we aim to help side-step this roadblock by exploring the usage of Natural Language Generation in augmenting datasets such that they can be used for NLP model development on downstream clinically relevant tasks. We propose a methodology guiding the generation with structured patient information in a sequence-to-sequence manner. We experiment with state-of-the-art Transformer models and demonstrate that our augmented dataset is capable of beating our baselines on a downstream classification task. Finally, we also create a user interface and release the scripts to train generation models to stimulate further research in this area.

pdf bib
Towards a Gold Standard for Evaluating Danish Word EmbeddingsDanish Word Embeddings
Nina Schneidermann | Rasmus Hvingelby | Bolette Pedersen

This paper presents the process of compiling a model-agnostic similarity goal standard for evaluating Danish word embeddings based on human judgments made by 42 native speakers of Danish. Word embeddings resemble semantic similarity solely by distribution (meaning that word vectors do not reflect relatedness as differing from similarity), and we argue that this generalization poses a problem in most intrinsic evaluation scenarios. In order to be able to evaluate on both dimensions, our human-generated dataset is therefore designed to reflect the distinction between relatedness and similarity. The goal standard is applied for evaluating the goodness of six existing word embedding models for Danish, and it is discussed how a relatively low correlation can be explained by the fact that semantic similarity is substantially more challenging to model than relatedness, and that there seems to be a need for future human judgments to measure similarity in full context and along more than a single spectrum.

pdf bib
Urban Dictionary Embeddings for Slang NLP ApplicationsNLP Applications
Steven Wilson | Walid Magdy | Barbara McGillivray | Kiran Garimella | Gareth Tyson

The choice of the corpus on which word embeddings are trained can have a sizable effect on the learned representations, the types of analyses that can be performed with them, and their utility as features for machine learning models. To contribute to the existing sets of pre-trained word embeddings, we introduce and release the first set of word embeddings trained on the content of Urban Dictionary, a crowd-sourced dictionary for slang words and phrases. We show that although these embeddings are trained on fewer total tokens (by at least an order of magnitude compared to most popular pre-trained embeddings), they have high performance across a range of common word embedding evaluations, ranging from semantic similarity to word clustering tasks. Further, for some extrinsic tasks such as sentiment analysis and sarcasm detection where we expect to require some knowledge of colloquial language on social media data, initializing classifiers with the Urban Dictionary Embeddings resulted in improved performance compared to initializing with a range of other well-known, pre-trained embeddings that are order of magnitude larger in size.

pdf bib
Aligning Wikipedia with WordNet : a Review and Evaluation of Different TechniquesWikipedia with WordNet:a Review and Evaluation of Different Techniques
Antoni Oliver

In this paper we explore techniques for aligning Wikipedia articles with WordNet synsets, their successful alignment being our main goal. We evaluate techniques that use the definitions and sense relations in Wordnet and the text and categories in Wikipedia articles. The results we present are based on two evaluation strategies : one uses a new gold and silver standard (for which the creation process is explained) ; the other creates wordnets in other languages and then compares them with existing wordnets for those languages found in the Open Multilingual Wordnet project. A reliable alignment between WordNet and Wikipedia is a very valuable resource for the creation of new wordnets in other languages and for the development of existing wordnets. The evaluation of alignments between WordNet and lexical resources is a difficult and time-consuming task, but the evaluation strategy using the Open Multilingual Wordnet can be used as an automated evaluation measure to assess the quality of alignments between these two resources.

pdf bib
The MWN.PT WordNet for Portuguese : Projection, Validation, Cross-lingual Alignment and DistributionMWN.PT WordNet for Portuguese: Projection, Validation, Cross-lingual Alignment and Distribution
António Branco | Sara Grilo | Márcia Bolrinha | Chakaveh Saedi | Ruben Branco | João Silva | Andreia Querido | Rita de Carvalho | Rosa Gaudio | Mariana Avelãs | Clara Pinto

The objective of the present paper is twofold, to present the MWN.PT WordNet and to report on its construction and on the lessons learned with it. The MWN.PT WordNet for Portuguese includes 41,000 concepts, expressed by 38,000 lexical units. Its synsets were manually validated and are linked to semantically equivalent synsets of the Princeton WordNet of English, and thus transitively to the many wordnets for other languages that are also linked to this English wordnet. To the best of our knowledge, it is the largest high quality, manually validated and cross-lingually integrated, wordnet of Portuguese distributed for reuse. Its construction was initiated more than one decade ago and its description is published for the first time in the present paper. It follows a three step projection, validation with alignment, completion methodology consisting on the manual validation and expansion of the outcome of an automatic projection procedure of synsets and their hypernym relations, followed by another automatic procedure that transferred the relations of remaining semantic types across wordnets of different languages.

pdf bib
The Ontology of Bulgarian Dialects Architecture and Information RetrievalBulgarian Dialects – Architecture and Information Retrieval
Rositsa Dekova

Following a concise description of the structure, the paper focuses on the potential of the Ontology of the Bulgarian Dialects, which demonstrates a novel usage of the ontological modelling for the purposes of dialect digital archiving and information processing. The ontology incorporates information on the dialects of the Bulgarian language and includes data from 84 dialects, spoken not only on the territory of the Republic of Bulgaria, but also abroad. It encodes both their geographical distribution and some of their main diagnostic features, such as the different mutations (also referred to as reflexes) of some of the Old Bulgarian vowels. The mutations modelled so far in the ontology include the reflex of the back nasal vowel // under stress, the reflex of the back er vowel // under stress, and the reflex of the yat vowel // under stress when it precedes a syllable with a back vowel. Besides the opportunity for formal structuring of the considerable amount of data gathered through the years by dialectologists, the ontology also provides numerous possibilities for information retrieval searches by dialect, country, dialect region, city or village, various combinations of diagnostic features.

pdf bib
Metaphorical Expressions in Automatic Arabic Sentiment AnalysisArabic Sentiment Analysis
Israa Alsiyat | Scott Piao

Over the recent years, Arabic language resources and NLP tools have been under rapid development. One of the important tasks for Arabic natural language processing is the sentiment analysis. While a significant improvement has been achieved in this research area, the existing computational models and tools still suffer from the lack of capability of dealing with Arabic metaphorical expressions. Metaphor has an important role in Arabic language due to its unique history and culture. Metaphors provide a linguistic mechanism for expressing ideas and notions that can be different from their surface form. Therefore, in order to efficiently identify true sentiment of Arabic language data, a computational model needs to be able to read between lines. In this paper, we examine the issue of metaphors in automatic Arabic sentiment analysis by carrying out an experiment, in which we observe the performance of a state-of-art Arabic sentiment tool on metaphors and analyse the result to gain a deeper insight into the issue. Our experiment evidently shows that metaphors have a significant impact on the performance of current Arabic sentiment tools, and it is an important task to develop Arabic language resources and computational models for Arabic metaphors.

pdf bib
Doctor Who? Framing Through Names and Titles in GermanGerman
Esther van den Berg | Katharina Korfhage | Josef Ruppenhofer | Michael Wiegand | Katja Markert

Entity framing is the selection of aspects of an entity to promote a particular viewpoint towards that entity. We investigate entity framing of political figures through the use of names and titles in German online discourse, enhancing current research in entity framing through titling and naming that concentrates on English only. We collect tweets that mention prominent German politicians and annotate them for stance. We find that the formality of naming in these tweets correlates positively with their stance. This confirms sociolinguistic observations that naming and titling can have a status-indicating function and suggests that this function is dominant in German tweets mentioning political figures. We also find that this status-indicating function is much weaker in tweets from users that are politically left-leaning than in tweets by right-leaning users. This is in line with observations from moral psychology that left-leaning and right-leaning users assign different importance to maintaining social hierarchies.

pdf bib
An Empirical Examination of Online Restaurant Reviews
Hyun Jung Kang | Iris Eshkol-Taravella

In the wake of (Pang et al., 2002 ; Turney, 2002 ; Liu, 2012) inter alia, opinion mining and sentiment analysis have focused on extracting either positive or negative opinions from texts and determining the targets of these opinions. In this study, we go beyond the coarse-grained positive vs. negative opposition and propose a corpus-based scheme that detects evaluative language at a finer-grained level. We classify each sentence into one of four evaluation types based on the proposed scheme : (1) the reviewer’s opinion on the restaurant (positive, negative, or mixed) ; (2) the reviewer’s input / feedback to potential customers and restaurant owners (suggestion, advice, or warning) (3) whether the reviewer wants to return to the restaurant (intention) ; (4) the factual statement about the experience (description). We apply classical machine learning and deep learning methods to show the effectiveness of our scheme. We also interpret the performances that we obtained for each category by taking into account the specificities of the corpus treated.

pdf bib
Annotating Perspectives on Vaccination
Roser Morante | Chantal van Son | Isa Maks | Piek Vossen

In this paper we present the Vaccination Corpus, a corpus of texts related to the online vaccination debate that has been annotated with three layers of information about perspectives : attribution, claims and opinions. Additionally, events related to the vaccination debate are also annotated. The corpus contains 294 documents from the Internet which reflect different views on vaccinations. It has been compiled to study the language of online debates, with the final goal of experimenting with methodologies to extract and contrast perspectives in the framework of the vaccination debate.

pdf bib
Dataset Creation and Evaluation of Aspect Based Sentiment Analysis in Telugu, a Low Resource LanguageTelugu, a Low Resource Language
Yashwanth Reddy Regatte | Rama Rohit Reddy Gangula | Radhika Mamidi

In recent years, sentiment analysis has gained popularity as it is essential to moderate and analyse the information across the internet. It has various applications like opinion mining, social media monitoring, and market research. Aspect Based Sentiment Analysis (ABSA) is an area of sentiment analysis which deals with sentiment at a finer level. ABSA classifies sentiment with respect to each aspect to gain greater insights into the sentiment expressed. Significant contributions have been made in ABSA, but this progress is limited only to a few languages with adequate resources. Telugu lags behind in this area of research despite being one of the most spoken languages in India and an enormous amount of data being created each day. In this paper, we create a reliable resource for aspect based sentiment analysis in Telugu. The data is annotated for three tasks namely Aspect Term Extraction, Aspect Polarity Classification and Aspect Categorisation. Further, we develop baselines for the tasks using deep learning methods demonstrating the reliability and usefulness of the resource.

pdf bib
A Fine-grained Sentiment Dataset for NorwegianNorwegian
Lilja Øvrelid | Petter Mæhlum | Jeremy Barnes | Erik Velldal

We here introduce NoReC_fine, a dataset for fine-grained sentiment analysis in Norwegian, annotated with respect to polar expressions, targets and holders of opinion. The underlying texts are taken from a corpus of professionally authored reviews from multiple news-sources and across a wide variety of domains, including literature, games, music, products, movies and more. We here present a detailed description of this annotation effort. We provide an overview of the developed annotation guidelines, illustrated with examples and present an analysis of inter-annotator agreement. We also report the first experimental results on the dataset, intended as a preliminary benchmark for further experiments.

pdf bib
The Design and Construction of a Chinese Sarcasm DatasetChinese Sarcasm Dataset
Xiaochang Gong | Qin Zhao | Jun Zhang | Ruibin Mao | Ruifeng Xu

As a typical multi-layered semi-conscious language phenomenon, sarcasm is widely existed in social media text for enhancing the emotion expression. Thus, the detection and processing of sarcasm is important to social media analysis. However, most existing sarcasm dataset are in English and there is still a lack of authoritative Chinese sarcasm dataset. In this paper, we presents the design and construction of a largest high-quality Chinese sarcasm dataset, which contains 2,486 manual annotated sarcastic texts and 89,296 non-sarcastic texts. Furthermore, a balanced dataset through elaborately sampling the same amount non-sarcastic texts for training sarcasm classifier. Using the dataset as the benchmark, some sarcasm classification methods are evaluated.

pdf bib
Target-based Sentiment Annotation in Chinese Financial NewsChinese Financial News
Chaofa Yuan | Yuhan Liu | Rongdi Yin | Jun Zhang | Qinling Zhu | Ruibin Mao | Ruifeng Xu

This paper presents the design and construction of a large-scale target-based sentiment annotation corpus on Chinese financial news text. Different from the most existing paragraph / document-based annotation corpus, in this study, target-based fine-grained sentiment annotation is performed. The companies, brands and other financial entities are regarded as the targets. The clause reflecting the profitability, loss or other business status of financial entities is regarded as the sentiment expression for determining the polarity. Based on high quality annotation guideline and effective quality control strategy, a corpus with 8,314 target-level sentiment annotation is constructed on 6,336 paragraphs from Chinese financial news text. Based on this corpus, several state-of-the-art sentiment analysis models are evaluated.

pdf bib
Reproduction and Revival of the Argument Reasoning Comprehension Task
João António Rodrigues | Ruben Branco | João Silva | António Branco

Reproduction of scientific findings is essential for scientific development across all scientific disciplines and reproducing results of previous works is a basic requirement for validating the hypothesis and conclusions put forward by them. This paper reports on the scientific reproduction of several systems addressing the Argument Reasoning Comprehension Task of SemEval2018. Given a recent publication that pointed out spurious statistical cues in the data set used in the shared task, and that produced a revised version of it, we also evaluated the reproduced systems with this new data set. The exercise reported here shows that, in general, the reproduction of these systems is successful with scores in line with those reported in SemEval2018. However, the performance scores are worst than those, and even below the random baseline, when the reproduced systems are run over the revised data set expunged from data artifacts. This demonstrates that this task is actually a much harder challenge than what could have been perceived from the inflated, close to human-level performance scores obtained with the data set used in SemEval2018. This calls for a revival of this task as there is much room for improvement until systems may come close to the upper bound provided by human performance.

pdf bib
ParlVote : A Corpus for Sentiment Analysis of Political DebatesParlVote: A Corpus for Sentiment Analysis of Political Debates
Gavin Abercrombie | Riza Batista-Navarro

Debate transcripts from the UK Parliament contain information about the positions taken by politicians towards important topics, but are difficult for people to process manually. While sentiment analysis of debate speeches could facilitate understanding of the speakers’ stated opinions, datasets currently available for this task are small when compared to the benchmark corpora in other domains. We present ParlVote, a new, larger corpus of parliamentary debate speeches for use in the evaluation of sentiment analysis systems for the political domain. We also perform a number of initial experiments on this dataset, testing a variety of approaches to the classification of sentiment polarity in debate speeches. These include a linear classifier as well as a neural network trained using a transformer word embedding model (BERT), and fine-tuned on the parliamentary speeches. We find that in many scenarios, a linear classifier trained on a bag-of-words text representation achieves the best results. However, with the largest dataset, the transformer-based model combined with a neural classifier provides the best performance. We suggest that further experimentation with classification models and observations of the debate content and structure are required, and that there remains much room for improvement in parliamentary sentiment analysis.

pdf bib
Offensive Language Identification in GreekGreek
Zesis Pitenis | Marcos Zampieri | Tharindu Ranasinghe

As offensive language has become a rising issue for online communities and social media platforms, researchers have been investigating ways of coping with abusive content and developing systems to detect its different types : cyberbullying, hate speech, aggression, etc. With a few notable exceptions, most research on this topic so far has dealt with English. This is mostly due to the availability of language resources for English. To address this shortcoming, this paper presents the first Greek annotated dataset for offensive language identification : the Offensive Greek Tweet Dataset (OGTD). OGTD is a manually annotated dataset containing 4,779 posts from Twitter annotated as offensive and not offensive. Along with a detailed description of the dataset, we evaluate several computational models trained and tested on this data.

pdf bib
GRAIN-S : Manually Annotated Syntax for German InterviewsGRAIN-S: Manually Annotated Syntax for German Interviews
Agnieszka Falenska | Zoltán Czesznak | Kerstin Jung | Moritz Völkel | Wolfgang Seeker | Jonas Kuhn

We present GRAIN-S, a set of manually created syntactic annotations for radio interviews in German. The dataset extends an existing corpus GRAIN and comes with constituency and dependency trees for six interviews. The rare combination of gold- and silver-standard annotation layers coming from GRAIN with high-quality syntax trees can serve as a useful resource for speech- and text-based research. Moreover, since interviews can be put between carefully prepared speech and spontaneous conversational speech, they cover phenomena not seen in traditional newspaper-based treebanks. Therefore, GRAIN-S can contribute to research into techniques for model adaptation and for building more corpus-independent tools. GRAIN-S follows TIGER, one of the established syntactic treebanks of German. We describe the annotation process and discuss decisions necessary to adapt the original TIGER guidelines to the interviews domain. Next, we give details on the conversion from TIGER-style trees to dependency trees. We provide data statistics and demonstrate differences between the new dataset and existing out-of-domain test sets annotated with TIGER syntactic structures. Finally, we provide baseline parsing results for further comparison.

pdf bib
The EDGeS Diachronic Bible CorpusEDGeS Diachronic Bible Corpus
Gerlof Bouma | Evie Coussé | Trude Dijkstra | Nicoline van der Sijs

We present the EDGeS Diachronic Bible Corpus : a diachronically and synchronically parallel corpus of Bible translations in Dutch, English, German and Swedish, with texts from the 14th century until today. It is compiled in the context of an intended longitudinal and contrastive study of complex verb constructions in Germanic. The paper discusses the corpus design principles, its selection of 36 Bibles, and the information and metadata encoded for the corpus texts. The EDGeS corpus will be available in two forms : the whole corpus will be accessible for researchers behind a login in the well-known OPUS search infrastructure, and the open subpart of the corpus will be available for download.

pdf bib
THEL : Automatically Extracted Typelogical Derivations for DutchÆTHEL: Automatically Extracted Typelogical Derivations for Dutch
Konstantinos Kogkalidis | Michael Moortgat | Richard Moot

We present THEL, a semantic compositionality dataset for written Dutch. THEL consists of two parts. First, it contains a lexicon of supertags for about 900 000 words in context. The supertags correspond to types of the simply typed linear lambda-calculus, enhanced with dependency decorations that capture grammatical roles supplementary to function-argument structures. On the basis of these types, THEL further provides 72 192 validated derivations, presented in four formats : natural-deduction and sequent-style proofs, linear logic proofnets and the associated programs (lambda terms) for meaning composition. THEL’s types and derivations are obtained by means of an extraction algorithm applied to the syntactic analyses of LASSY Small, the gold standard corpus of written Dutch. We discuss the extraction algorithm and show how ‘virtual elements’ in the original LASSY annotation of unbounded dependencies and coordination phenomena give rise to higher-order types. We suggest some example usecases highlighting the benefits of a type-driven approach at the syntax semantics interface. The following resources are open-sourced with THEL : the lexical mappings between words and types, a subset of the dataset consisting of 7 924 semantic parses, and the Python code that implements the extraction algorithm.

pdf bib
AlloVera : A Multilingual Allophone DatabaseAlloVera: A Multilingual Allophone Database
David R. Mortensen | Xinjian Li | Patrick Littell | Alexis Michaud | Shruti Rijhwani | Antonios Anastasopoulos | Alan W Black | Florian Metze | Graham Neubig

We introduce a new resource, AlloVera, which provides mappings from 218 allophones to phonemes for 14 languages. Phonemes are contrastive phonological units, and allophones are their various concrete realizations, which are predictable from phonological context. While phonemic representations are language specific, phonetic representations (stated in terms of (allo)phones) are much closer to a universal (language-independent) transcription. AlloVera allows the training of speech recognition models that output phonetic transcriptions in the International Phonetic Alphabet (IPA), regardless of the input language. We show that a universal allophone model, Allosaurus, built with AlloVera, outperforms universal phonemic models and language-specific models on a speech-transcription task. We explore the implications of this technology (and related technologies) for the documentation of endangered and minority languages. We further explore other applications for which AlloVera will be suitable as it grows, including phonological typology.

pdf bib
AccentDB : A Database of Non-Native English Accents to Assist Neural Speech RecognitionAccentDB: A Database of Non-Native English Accents to Assist Neural Speech Recognition
Afroz Ahamad | Ankit Anand | Pranesh Bhargava

Modern Automatic Speech Recognition (ASR) technology has evolved to identify the speech spoken by native speakers of a language very well. However, identification of the speech spoken by non-native speakers continues to be a major challenge for it. In this work, we first spell out the key requirements for creating a well-curated database of speech samples in non-native accents for training and testing robust ASR systems. We then introduce AccentDB, one such database that contains samples of 4 Indian-English accents collected by us, and a compilation of samples from 4 native-English, and a metropolitan Indian-English accent. We also present an analysis on separability of the collected accent data. Further, we present several accent classification models and evaluate them thoroughly against human-labelled accent classes. We test the generalization of our classifier models in a variety of setups of seen and unseen data. Finally, we introduce accent neutralization of non-native accents to native accents using autoencoder models with task-specific architectures. Thus, our work aims to aid ASR systems at every stage of development with a database for training, classification models for feature augmentation, and neutralization systems for acoustic transformations of non-native accents of English.

pdf bib
Propagate-Selector : Detecting Supporting Sentences for Question Answering via Graph Neural Networks
Seunghyun Yoon | Franck Dernoncourt | Doo Soon Kim | Trung Bui | Kyomin Jung

In this study, we propose a novel graph neural network called propagate-selector (PS), which propagates information over sentences to understand information that can not be inferred when considering sentences in isolation. First, we design a graph structure in which each node represents an individual sentence, and some pairs of nodes are selectively connected based on the text structure. Then, we develop an iterative attentive aggregation and a skip-combine method in which a node interacts with its neighborhood nodes to accumulate the necessary information. To evaluate the performance of the proposed approaches, we conduct experiments with the standard HotpotQA dataset. The empirical results demonstrate the superiority of our proposed approach, which obtains the best performances, compared to the widely used answer-selection models that do not consider the intersentential relationship.

pdf bib
Cross-sentence Pre-trained Model for Interactive QA matchingQA matching
Jinmeng Wu | Yanbin Hao

Semantic matching measures the dependencies between query and answer representations, it is an important criterion for evaluating whether the matching is successful. In fact, such matching does not examine each sentence individually, context information outside a sentence should be considered equally important to the syntactic context inside a sentence. We proposed a new QA matching model, built upon a cross-sentence context-aware architecture. An interactive attention mechanism with a pre-trained language model is proposed to automatically select salient positional answer representations that contribute more significantly to the answer relevance of a given question. In addition to the context information captured at each word position, we incorporate a new quantity of context information jump to facilitate the attention weight formulation. This reflects the amount of new information brought by the next word and is computed by modeling the joint probability between two adjacent word states. The proposed method is compared to multiple state-of-the-art ones evaluated using the TREC library, WikiQA, and the Yahoo ! community question datasets. Experimental results show that the proposed method outperforms satisfactorily the competing ones.

pdf bib
SQuAD2-CR : Semi-supervised Annotation for Cause and Rationales for Unanswerability in SQuAD 2.0SQuAD2-CR: Semi-supervised Annotation for Cause and Rationales for Unanswerability in SQuAD 2.0
Gyeongbok Lee | Seung-won Hwang | Hyunsouk Cho

Existing machine reading comprehension models are reported to be brittle for adversarially perturbed questions when optimizing only for accuracy, which led to the creation of new reading comprehension benchmarks, such as SQuAD 2.0 which contains such type of questions. However, despite the super-human accuracy of existing models on such datasets, it is still unclear how the model predicts the answerability of the question, potentially due to the absence of a shared annotation for the explanation. To address such absence, we release SQuAD2-CR dataset, which contains annotations on unanswerable questions from the SQuAD 2.0 dataset, to enable an explanatory analysis of the model prediction. Specifically, we annotate (1) explanation on why the most plausible answer span can not be the answer and (2) which part of the question causes unanswerability. We share intuitions and experimental results that how this dataset can be used to analyze and improve the interpretability of existing reading comprehension model behavior.

pdf bib
Generating Responses that Reflect Meta Information in User-Generated Question Answer Pairs
Takashi Kodama | Ryuichiro Higashinaka | Koh Mitsuda | Ryo Masumura | Yushi Aono | Ryuta Nakamura | Noritake Adachi | Hidetoshi Kawabata

This paper concerns the problem of realizing consistent personalities in neural conversational modeling by using user generated question-answer pairs as training data. Using the framework of role play-based question answering, we collected single-turn question-answer pairs for particular characters from online users. Meta information was also collected such as emotion and intimacy related to question-answer pairs. We verified the quality of the collected data and, by subjective evaluation, we also verified their usefulness in training neural conversational models for generating utterances reflecting the meta information, especially emotion.

pdf bib
KGvec2go Knowledge Graph Embeddings as a ServiceKGvec2go – Knowledge Graph Embeddings as a Service
Jan Portisch | Michael Hladik | Heiko Paulheim

In this paper, we present KGvec2go, a Web API for accessing and consuming graph embeddings in a light-weight fashion in downstream applications. Currently, we serve pre-trained embeddings for four knowledge graphs. We introduce the service and its usage, and we show further that the trained models have semantic value by evaluating them on multiple semantic benchmarks. The evaluation also reveals that the combination of multiple models can lead to a better outcome than the best individual model.

pdf bib
Defying Wikidata : Validation of Terminological Relations in the Web of DataWikidata: Validation of Terminological Relations in the Web of Data
Patricia Martín-Chozas | Sina Ahmadi | Elena Montiel-Ponsoda

In this paper we present an approach to validate terminological data retrieved from open encyclopaedic knowledge bases. This need arises from the enrichment of automatically extracted terms with information from existing resources in theLinguistic Linked Open Data cloud. Specifically, the resource employed for this enrichment is WIKIDATA, since it is one of the biggest knowledge bases freely available within the Semantic Web. During the experiment, we noticed that certain RDF properties in the Knowledge Base did not contain the data they are intended to represent, but a different type of information. In this paper we propose an approach to validate the retrieved data based on four axioms that rely on two linguistic theories : the x-bar theory and the multidimensional theory of terminology. The validation process is supported by a second knowledge base specialised in linguistic data ; in this case, CONCEPTNET. In our experiment, we validate terms from the legal domain in four languages : Dutch, English, German and Spanish. The final aim is to generate a set of sound and reliable terminological resources in RDF to contribute to the population of the Linguistic Linked Open Data cloud.

pdf bib
The Universal Decompositional Semantics Dataset and Decomp Toolkit
Aaron Steven White | Elias Stengel-Eskin | Siddharth Vashishtha | Venkata Subrahmanyan Govindarajan | Dee Ann Reisinger | Tim Vieira | Keisuke Sakaguchi | Sheng Zhang | Francis Ferraro | Rachel Rudinger | Kyle Rawlins | Benjamin Van Durme

We present the Universal Decompositional Semantics (UDS) dataset (v1.0), which is bundled with the Decomp toolkit (v0.1). UDS1.0 unifies five high-quality, decompositional semantics-aligned annotation sets within a single semantic graph specificationwith graph structures defined by the predicative patterns produced by the PredPatt tool and real-valued node and edge attributes constructed using sophisticated normalization procedures. The Decomp toolkit provides a suite of Python 3 tools for querying UDS graphs using SPARQL. Both UDS1.0 and Decomp0.1 are publicly available at

pdf bib
Are Word Embeddings Really a Bad Fit for the Estimation of Thematic Fit?
Emmanuele Chersoni | Ludovica Pannitto | Enrico Santus | Alessandro Lenci | Chu-Ren Huang

While neural embeddings represent a popular choice for word representation in a wide variety of NLP tasks, their usage for thematic fit modeling has been limited, as they have been reported to lag behind syntax-based count models. In this paper, we propose a complete evaluation of count models and word embeddings on thematic fit estimation, by taking into account a larger number of parameters and verb roles and introducing also dependency-based embeddings in the comparison. Our results show a complex scenario, where a determinant factor for the performance seems to be the availability to the model of reliable syntactic information for building the distributional representations of the roles.

pdf bib
Ciron : a New Benchmark Dataset for Chinese Irony DetectionCiron: a New Benchmark Dataset for Chinese Irony Detection
Rong Xiang | Xuefeng Gao | Yunfei Long | Anran Li | Emmanuele Chersoni | Qin Lu | Chu-Ren Huang

Automatic Chinese irony detection is a challenging task, and it has a strong impact on linguistic research. However, Chinese irony detection often lacks labeled benchmark datasets. In this paper, we introduce Ciron, the first Chinese benchmark dataset available for irony detection for machine learning models. Ciron includes more than 8.7 K posts, collected from Weibo, a micro blogging platform. Most importantly, Ciron is collected with no pre-conditions to ensure a much wider coverage. Evaluation on seven different machine learning classifiers proves the usefulness of Ciron as an important resource for Chinese irony detection.

pdf bib
NegBERT : A Transfer Learning Approach for Negation Detection and Scope ResolutionNegBERT: A Transfer Learning Approach for Negation Detection and Scope Resolution
Aditya Khandelwal | Suraj Sawant

Negation is an important characteristic of language, and a major component of information extraction from text. This subtask is of considerable importance to the biomedical domain. Over the years, multiple approaches have been explored to address this problem : Rule-based systems, Machine Learning classifiers, Conditional Random Field models, CNNs and more recently BiLSTMs. In this paper, we look at applying Transfer Learning to this problem. First, we extensively review previous literature addressing Negation Detection and Scope Resolution across the 3 datasets that have gained popularity over the years : the BioScope Corpus, the Sherlock dataset, and the SFU Review Corpus. We then explore the decision choices involved with using BERT, a popular transfer learning model, for this task, and report state-of-the-art results for scope resolution across all 3 datasets. Our model, referred to as NegBERT, achieves a token level F1 score on scope resolution of 92.36 on the Sherlock dataset, 95.68 on the BioScope Abstracts subcorpus, 91.24 on the BioScope Full Papers subcorpus, 90.95 on the SFU Review Corpus, outperforming the previous state-of-the-art systems by a significant margin. We also analyze the model’s generalizability to datasets on which it is not trained.

pdf bib
Spatial Multi-Arrangement for Clustering and Multi-way Similarity Dataset Construction
Olga Majewska | Diana McCarthy | Jasper van den Bosch | Nikolaus Kriegeskorte | Ivan Vulić | Anna Korhonen

We present a novel methodology for fast bottom-up creation of large-scale semantic similarity resources to support development and evaluation of NLP systems. Our work targets verb similarity, but the methodology is equally applicable to other parts of speech. Our approach circumvents the bottleneck of slow and expensive manual development of lexical resources by leveraging semantic intuitions of native speakers and adapting a spatial multi-arrangement approach from cognitive neuroscience, used before only with visual stimuli, to lexical stimuli. Our approach critically obtains judgments of word similarity in the context of a set of related words, rather than of word pairs in isolation. We also handle lexical ambiguity as a natural consequence of a two-phase process where verbs are placed in broad semantic classes prior to the fine-grained spatial similarity judgments. Our proposed design produces a large-scale verb resource comprising 17 relatedness-based classes and a verb similarity dataset containing similarity scores for 29,721 unique verb pairs and 825 target verbs, which we release with this paper.

pdf bib
A Short Survey on Sense-Annotated Corpora
Tommaso Pasini | Jose Camacho-Collados

Large sense-annotated datasets are increasingly necessary for training deep supervised systems in Word Sense Disambiguation. However, gathering high-quality sense-annotated data for as many instances as possible is a laborious and expensive task. This has led to the proliferation of automatic and semi-automatic methods for overcoming the so-called knowledge-acquisition bottleneck. In this short survey we present an overview of sense-annotated corpora, annotated either manually- or (semi)automatically, that are currently available for different languages and featuring distinct lexical resources as inventory of senses, i.e. WordNet, Wikipedia, BabelNet. Furthermore, we provide the reader with general statistics of each dataset and an analysis of their specific features.

pdf bib
NUBes : A Corpus of Negation and Uncertainty in Spanish Clinical TextsNUBes: A Corpus of Negation and Uncertainty in Spanish Clinical Texts
Salvador Lima Lopez | Naiara Perez | Montse Cuadros | German Rigau

This paper introduces the first version of the NUBes corpus (Negation and Uncertainty annotations in Biomedical texts in Spanish). The corpus is part of an on-going research and currently consists of 29,682 sentences obtained from anonymised health records annotated with negation and uncertainty. The article includes an exhaustive comparison with similar corpora in Spanish, and presents the main annotation and design decisions. Additionally, we perform preliminary experiments using deep learning algorithms to validate the annotated dataset. As far as we know, NUBes is the largest available corpora for negation in Spanish and the first that also incorporates the annotation of speculation cues, scopes, and events.

pdf bib
Decomposing and Comparing Meaning Relations : Paraphrasing, Textual Entailment, Contradiction, and Specificity
Venelin Kovatchev | Darina Gold | M. Antonia Marti | Maria Salamo | Torsten Zesch

In this paper, we present a methodology for decomposing and comparing multiple meaning relations (paraphrasing, textual entailment, contradiction, and specificity). The methodology includes SHARel-a new typology that consists of 26 linguistic and 8 reason-based categories. We use the typology to annotate a corpus of 520 sentence pairs in English and we demonstrate that unlike previous typologies, SHARel can be applied to all relations of interest with a high inter-annotator agreement. We analyze and compare the frequency and distribution of the linguistic and reason-based phenomena involved in paraphrasing, textual entailment, contradiction, and specificity. This comparison allows for a much more in-depth analysis of the workings of the individual relations and the way they interact and compare with each other. We release all resources (typology, annotation guidelines, and annotated corpus) to the community.

pdf bib
Figure Me Out : A Gold Standard Dataset for Metaphor Interpretation
Omnia Zayed | John Philip McCrae | Paul Buitelaar

Metaphor comprehension and understanding is a complex cognitive task that requires interpreting metaphors by grasping the interaction between the meaning of their target and source concepts. This is very challenging for humans, let alone computers. Thus, automatic metaphor interpretation is understudied in part due to the lack of publicly available datasets. The creation and manual annotation of such datasets is a demanding task which requires huge cognitive effort and time. Moreover, there will always be a question of accuracy and consistency of the annotated data due to the subjective nature of the problem. This work addresses these issues by presenting an annotation scheme to interpret verb-noun metaphoric expressions in text. The proposed approach is designed with the goal of reducing the workload on annotators and maintain consistency. Our methodology employs an automatic retrieval approach which utilises external lexical resources, word embeddings and semantic similarity to generate possible interpretations of identified metaphors in order to enable quick and accurate annotation. We validate our proposed approach by annotating around 1,500 metaphors in tweets which were annotated by six native English speakers. As a result of this work, we publish as linked data the first gold standard dataset for metaphor interpretation which will facilitate research in this area.

pdf bib
Dataset and Enhanced Model for Eligibility Criteria-to-SQL Semantic ParsingSQL Semantic Parsing
Xiaojing Yu | Tianlong Chen | Zhengjie Yu | Huiyu Li | Yang Yang | Xiaoqian Jiang | Anxiao Jiang

Clinical trials often require that patients meet eligibility criteria (e.g., have specific conditions) to ensure the safety and the effectiveness of studies. However, retrieving eligible patients for a trial from the electronic health record (EHR) database remains a challenging task for clinicians since it requires not only medical knowledge about eligibility criteria, but also an adequate understanding of structured query language (SQL). In this paper, we introduce a new dataset that includes the first-of-its-kind eligibility-criteria corpus and the corresponding queries for criteria-to-sql (Criteria2SQL), a task translating the eligibility criteria to executable SQL queries. Compared to existing datasets, the queries in the dataset here are derived from the eligibility criteria of clinical trials and include Order-sensitive, Counting-based, and Boolean-type cases which are not seen before. In addition to the dataset, we propose a novel neural semantic parser as a strong baseline model. Extensive experiments show that the proposed parser outperforms existing state-of-the-art general-purpose text-to-sql models while highlighting the challenges presented by the new dataset. The uniqueness and the diversity of the dataset leave a lot of research opportunities for future improvement.Order-sensitive, Counting-based, and Boolean-type cases which are not seen before. In addition to the dataset, we propose a novel neural semantic parser as a strong baseline model. Extensive experiments show that the proposed parser outperforms existing state-of-the-art general-purpose text-to-sql models while highlighting the challenges presented by the new dataset. The uniqueness and the diversity of the dataset leave a lot of research opportunities for future improvement.

pdf bib
Word Attribute Prediction Enhanced by Lexical Entailment Tasks
Mika Hasegawa | Tetsunori Kobayashi | Yoshihiko Hayashi

Human semantic knowledge about concepts acquired through perceptual inputs and daily experiences can be expressed as a bundle of attributes. Unlike the conventional distributed word representations that are purely induced from a text corpus, a semantic attribute is associated with a designated dimension in attribute-based vector representations. Thus, semantic attribute vectors can effectively capture the commonalities and differences among concepts. However, as semantic attributes have been generally created by psychological experimental settings involving human annotators, an automatic method to create or extend such resources is highly demanded in terms of language resource development and maintenance. This study proposes a two-stage neural network architecture, Word2Attr, in which initially acquired attribute representations are then fine-tuned by employing supervised lexical entailment tasks. The quantitative empirical results demonstrated that the fine-tuning was indeed effective in improving the performances of semantic / visual similarity / relatedness evaluation tasks. Although the qualitative analysis confirmed that the proposed method could often discover valid but not-yet human-annotated attributes, they also exposed future issues to be worked : we should refine the inventory of semantic attributes that currently relies on an existing dataset.

pdf bib
Representing Verbs with Visual Argument Vectors
Irene Sucameli | Alessandro Lenci

Is it possible to use images to model verb semantic similarities? Starting from this core question, we developed two textual distributional semantic models and a visual one. We found particularly interesting and challenging to investigate this Part of Speech since verbs are not often analysed in researches focused on multimodal distributional semantics. After the creation of the visual and textual distributional space, the three models were evaluated in relation to SimLex-999, a gold standard resource. Through this evaluation, we demonstrate that, using visual distributional models, it is possible to extract meaningful information and to effectively capture the semantic similarity between verbs.

pdf bib
A French Version of the FraCaS Test SuiteFrench Version of the FraCaS Test Suite
Maxime Amblard | Clément Beysson | Philippe de Groote | Bruno Guillaume | Sylvain Pogodalla

This paper presents a French version of the FraCaS test suite. This test suite, originally written in English, contains problems illustrating semantic inference in natural language. We describe linguistic choices we had to make when translating the FraCaS test suite in French, and discuss some of the issues that were raised by the translation. We also report an experiment we ran in order to test both the translation and the logical semantics underlying the problems of the test suite. This provides a way of checking formal semanticists’ hypotheses against actual semantic capacity of speakers (in the present case, French speakers), and allow us to compare the results we obtained with the ones of similar experiments that have been conducted for other languages.

pdf bib
Word Sense Disambiguation for 158 Languages using Word Embeddings Only
Varvara Logacheva | Denis Teslenko | Artem Shelmanov | Steffen Remus | Dmitry Ustalov | Andrey Kutuzov | Ekaterina Artemova | Chris Biemann | Simone Paolo Ponzetto | Alexander Panchenko

Disambiguation of word senses in context is easy for humans, but is a major challenge for automatic approaches. Sophisticated supervised and knowledge-based models were developed to solve this task. However, (i) the inherent Zipfian distribution of supervised training instances for a given word and/or (ii) the quality of linguistic knowledge representations motivate the development of completely unsupervised and knowledge-free approaches to word sense disambiguation (WSD). They are particularly useful for under-resourced languages which do not have any resources for building either supervised and/or knowledge-based models. In this paper, we present a method that takes as input a standard pre-trained word embedding model and induces a fully-fledged word sense inventory, which can be used for disambiguation in context. We use this method to induce a collection of sense inventories for 158 languages on the basis of the original pre-trained fastText word embeddings by Grave et al., (2018), enabling WSD in these languages. Models and system are available online.

pdf bib
One Classifier for All Ambiguous Words : Overcoming Data Sparsity by Utilizing Sense Correlations Across Words
Prafulla Kumar Choubey | Ruihong Huang

Most supervised word sense disambiguation (WSD) systems build word-specific classifiers by leveraging labeled data. However, when using word-specific classifiers, the sparseness of annotations leads to inferior sense disambiguation performance on less frequently seen words. To combat data sparsity, we propose to learn a single model that derives sense representations and meanwhile enforces congruence between a word instance and its right sense by using both sense-annotated data and lexical resources. The model is shared across words that allows utilizing sense correlations across words, and therefore helps to transfer common disambiguation rules from annotation-rich words to annotation-lean words. Empirical evaluation on benchmark datasets shows that the proposed shared model outperforms the equivalent classifier-based models by 1.7 %, 2.5 % and 3.8 % in F1-score when using GloVe, ELMo and BERT word embeddings respectively.

pdf bib
What Comes First : Combining Motion Capture and Eye Tracking Data to Study the Order of Articulators in Constructed Action in Sign Language Narratives
Tommi Jantunen | Anna Puupponen | Birgitta Burger

We use synchronized 120 fps motion capture and 50 fps eye tracking data from two native signers to investigate the temporal order in which the dominant hand, the head, the chest and the eyes start producing overt constructed action from regular narration in seven short Finnish Sign Language stories. From the material, we derive a sample of ten instances of regular narration to overt constructed action transfers in ELAN which we then further process and analyze in Matlab. The results indicate that the temporal order of articulators shows both contextual and individual variation but that there are also repeated patterns which are similar across all the analyzed sequences and signers. Most notably, when the discourse strategy changes from regular narration to overt constructed action, the head and the eyes tend to take the leading role, and the chest and the dominant hand tend to start acting last. Consequences of the findings are discussed.

pdf bib
LSF-ANIMAL : A Motion Capture Corpus in French Sign Language Designed for the Animation of Signing AvatarsLSF-ANIMAL: A Motion Capture Corpus in French Sign Language Designed for the Animation of Signing Avatars
Lucie Naert | Caroline Larboulette | Sylvie Gibet

Signing avatars allow deaf people to access information in their preferred language using an interactive visualization of the sign language spatio-temporal content. However, avatars are often procedurally animated, resulting in robotic and unnatural movements, which are therefore rejected by the community for which they are intended. To overcome this lack of authenticity, solutions in which the avatar is animated from motion capture data are promising. Yet, the initial data set drastically limits the range of signs that the avatar can produce. Therefore, it can be interesting to enrich the initial corpus with new content by editing the captured motions. For this purpose, we collected the LSF-ANIMAL corpus, a French Sign Language (LSF) corpus composed of captured isolated signs and full sentences that can be used both to study LSF features and to generate new signs and utterances. This paper presents the precise definition and content of this corpus, technical considerations relative to the motion capture process (including the marker set definition), the post-processing steps required to obtain data in a standard motion format and the annotation scheme used to label the data. The quality of the corpus with respect to intelligibility, accuracy and realism is perceptually evaluated by 41 participants including native LSF signers.

pdf bib
Sign Language Recognition with Transformer Networks
Mathieu De Coster | Mieke Van Herreweghe | Joni Dambre

Sign languages are complex languages. Research into them is ongoing, supported by large video corpora of which only small parts are annotated. Sign language recognition can be used to speed up the annotation process of these corpora, in order to aid research into sign languages and sign language recognition. Previous research has approached sign language recognition in various ways, using feature extraction techniques or end-to-end deep learning. In this work, we apply a combination of feature extraction using OpenPose for human keypoint estimation and end-to-end feature learning with Convolutional Neural Networks. The proven multi-head attention mechanism used in transformers is applied to recognize isolated signs in the Flemish Sign Language corpus. Our proposed method significantly outperforms the previous state of the art of sign language recognition on the Flemish Sign Language corpus : we obtain an accuracy of 74.7 % on a vocabulary of 100 classes. Our results will be implemented as a suggestion system for sign language corpus annotation.

pdf bib
Dicta-Sign-LSF-v2 : Remake of a Continuous French Sign Language Dialogue Corpus and a First Baseline for Automatic Sign Language ProcessingDicta-Sign-LSF-v2: Remake of a Continuous French Sign Language Dialogue Corpus and a First Baseline for Automatic Sign Language Processing
Valentin Belissen | Annelies Braffort | Michèle Gouiffès

While the research in automatic Sign Language Processing (SLP) is growing, it has been almost exclusively focused on recognizing lexical signs, whether isolated or within continuous SL production. However, Sign Languages include many other gestural units like iconic structures, which need to be recognized in order to go towards a true SL understanding. In this paper, we propose a newer version of the publicly available SL corpus Dicta-Sign, limited to its French Sign Language part. Involving 16 different signers, this dialogue corpus was produced with very few constraints on the style and content. It includes lexical and non-lexical annotations over 11 hours of video recording, with 35000 manual units. With the aim of stimulating research in SL understanding, we also provide a baseline for the recognition of lexical signs and non-lexical structures on this corpus. A very compact modeling of a signer is built and a Convolutional-Recurrent Neural Network is trained and tested on Dicta-Sign-LSF-v2, with state-of-the-art results, including the ability to detect iconicity in SL production.

pdf bib
Evaluation of Manual and Non-manual Components for Sign Language Recognition
Medet Mukushev | Arman Sabyrov | Alfarabi Imashev | Kenessary Koishybay | Vadim Kimmelman | Anara Sandygulova

The motivation behind this work lies in the need to differentiate between similar signs that differ in non-manual components present in any sign. To this end, we recorded full sentences signed by five native signers and extracted 5200 isolated sign samples of twenty frequently used signs in Kazakh-Russian Sign Language (K-RSL), which have similar manual components but differ in non-manual components (i.e. facial expressions, eyebrow height, mouth, and head orientation). We conducted a series of evaluations in order to investigate whether non-manual components would improve sign’s recognition accuracy. Among standard machine learning approaches, Logistic Regression produced the best results, 78.2 % of accuracy for dataset with 20 signs and 77.9 % of accuracy for dataset with 2 classes (statement vs question). Dataset can be downloaded from the following website :

pdf bib
A Survey on Natural Language Processing for Fake News Detection
Ray Oshikawa | Jing Qian | William Yang Wang

Fake news detection is a critical yet challenging problem in Natural Language Processing (NLP). The rapid rise of social networking platforms has not only yielded a vast increase in information accessibility but has also accelerated the spread of fake news. Thus, the effect of fake news has been growing, sometimes extending to the offline world and threatening public safety. Given the massive amount of Web content, automatic fake news detection is a practical NLP problem useful to all online content providers, in order to reduce the human time and effort to detect and prevent the spread of fake news. In this paper, we describe the challenges involved in fake news detection and also describe related tasks. We systematically review and compare the task formulations, datasets and NLP solutions that have been developed for this task, and also discuss the potentials and limitations of them. Based on our insights, we outline promising research directions, including more fine-grained, detailed, fair, and practical detection models. We also highlight the difference between fake news detection and other related tasks, and the importance of NLP solutions for fake news detection.

pdf bib
RP-DNN : A Tweet Level Propagation Context Based Deep Neural Networks for Early Rumor Detection in Social MediaRP-DNN: A Tweet Level Propagation Context Based Deep Neural Networks for Early Rumor Detection in Social Media
Jie Gao | Sooji Han | Xingyi Song | Fabio Ciravegna

Early rumor detection (ERD) on social media platform is very challenging when limited, incomplete and noisy information is available. Most of the existing methods have largely worked on event-level detection that requires the collection of posts relevant to a specific event and relied only on user-generated content. They are not appropriate to detect rumor sources in the very early stages, before an event unfolds and becomes widespread. In this paper, we address the task of ERD at the message level. We present a novel hybrid neural network architecture, which combines a task-specific character-based bidirectional language model and stacked Long Short-Term Memory (LSTM) networks to represent textual contents and social-temporal contexts of input source tweets, for modelling propagation patterns of rumors in the early stages of their development. We apply multi-layered attention models to jointly learn attentive context embeddings over multiple context inputs. Our experiments employ a stringent leave-one-out cross-validation (LOO-CV) evaluation setup on seven publicly available real-life rumor event data sets. Our models achieve state-of-the-art(SoA) performance for detecting unseen rumors on large augmented data which covers more than 12 events and 2,967 rumors. An ablation study is conducted to understand the relative contribution of each component of our proposed model.

pdf bib
Fakeddit : A New Multimodal Benchmark Dataset for Fine-grained Fake News DetectionFakeddit: A New Multimodal Benchmark Dataset for Fine-grained Fake News Detection
Kai Nakamura | Sharon Levy | William Yang Wang

Fake news has altered society in negative ways in politics and culture. It has adversely affected both online social network systems as well as offline communities and conversations. Using automatic machine learning classification models is an efficient way to combat the widespread dissemination of fake news. However, a lack of effective, comprehensive datasets has been a problem for fake news research and detection model development. Prior fake news datasets do not provide multimodal text and image data, metadata, comment data, and fine-grained fake news categorization at the scale and breadth of our dataset. We present Fakeddit, a novel multimodal dataset consisting of over 1 million samples from multiple categories of fake news. After being processed through several stages of review, the samples are labeled according to 2-way, 3-way, and 6-way classification categories through distant supervision. We construct hybrid text+image models and perform extensive experiments for multiple variations of classification, demonstrating the importance of the novel aspect of multimodality and fine-grained classification unique to Fakeddit.

pdf bib
I Feel Offended, Do n’t Be Abusive ! Implicit / Explicit Messages in Offensive and Abusive LanguageI Feel Offended, Don’t Be Abusive! Implicit/Explicit Messages in Offensive and Abusive Language
Tommaso Caselli | Valerio Basile | Jelena Mitrović | Inga Kartoziya | Michael Granitzer

Abusive language detection is an unsolved and challenging problem for the NLP community. Recent literature suggests various approaches to distinguish between different language phenomena (e.g., hate speech vs. cyberbullying vs. offensive language) and factors (degree of explicitness and target) that may help to classify different abusive language phenomena. There are data sets that annotate the target of abusive messages (i.e. OLID / OffensEval (Zampieri et al., 2019a)). However, there is a lack of data sets that take into account the degree of explicitness. In this paper, we propose annotation guidelines to distinguish between explicit and implicit abuse in English and apply them to OLID / OffensEval. The outcome is a newly created resource, AbuseEval v1.0, which aims to address some of the existing issues in the annotation of offensive and abusive language (e.g., explicitness of the message, presence of a target, need of context, and interaction across different phenomena).

pdf bib
Detecting Troll Tweets in a Bilingual Corpus
Lin Miao | Mark Last | Marina Litvak

During the past several years, a large amount of troll accounts has emerged with efforts to manipulate public opinion on social network sites. They are often involved in spreading misinformation, fake news, and propaganda with the intent of distracting and sowing discord. This paper aims to detect troll tweets in both English and Russian assuming that the tweets are generated by some troll farm. We reduce this task to the authorship verification problem of determining whether a single tweet is authored by a troll farm account or not. We evaluate a supervised classification approach with monolingual, cross-lingual, and bilingual training scenarios, using several machine learning algorithms, including deep learning. The best results are attained by the bilingual learning, showing the area under the ROC curve (AUC) of 0.875 and 0.828, for tweet classification in English and Russian test sets, respectively. It is noteworthy that these results are obtained using only raw text features, which do not require manual feature engineering efforts. In this paper, we introduce a resource of English and Russian troll tweets containing original tweets and translation from English to Russian, Russian to English. It is available for academic purposes.

pdf bib
Norm It ! Lexical Normalization for Italian and Its Downstream Effects for Dependency ParsingItalian and Its Downstream Effects for Dependency Parsing
Rob van der Goot | Alan Ramponi | Tommaso Caselli | Michele Cafagna | Lorenzo De Mattei

Lexical normalization is the task of translating non-standard social media data to a standard form. Previous work has shown that this is beneficial for many downstream tasks in multiple languages. However, for Italian, there is no benchmark available for lexical normalization, despite the presence of many benchmarks for other tasks involving social media data. In this paper, we discuss the creation of a lexical normalization dataset for Italian. After two rounds of annotation, a Cohen’s kappa score of 78.64 is obtained. During this process, we also analyze the inter-annotator agreement for this task, which is only rarely done on datasets for lexical normalization, and when it is reported, the analysis usually remains shallow. Furthermore, we utilize this dataset to train a lexical normalization model and show that it can be used to improve dependency parsing of social media data. All annotated data and the code to reproduce the results are available at :

pdf bib
Large Corpus of Czech Parliament Plenary HearingsCzech Parliament Plenary Hearings
Jonas Kratochvil | Peter Polak | Ondrej Bojar

We present a large corpus of Czech parliament plenary sessions. The corpus consists of approximately 1200 hours of speech data and corresponding text transcriptions. The whole corpus has been segmented to short audio segments making it suitable for both training and evaluation of automatic speech recognition (ASR) systems. The source language of the corpus is Czech, which makes it a valuable resource for future research as only a few public datasets are available in the Czech language. We complement the data release with experiments of two baseline ASR systems trained on the presented data : the more traditional approach implemented in the Kaldi ASRtoolkit which combines hidden Markov models and deep neural networks (NN) and a modern ASR architecture implemented in Jaspertoolkit which uses deep NNs in an end-to-end fashion.

pdf bib
Automatic Period Segmentation of Oral FrenchFrench
Natalia Kalashnikova | Loïc Grobol | Iris Eshkol-Taravella | François Delafontaine

Natural Language Processing in oral speech segmentation is still looking for a minimal unit to analyze. In this work, we present a comparison of two automatic segmentation methods of macro-syntactic periods which allows to take into account syntactic and prosodic components of speech. We compare the performances of an existing tool Analor (Avanzi, Lacheret-Dujour, Victorri, 2008) developed for automatic segmentation of prosodic periods and of CRF models relying on syntactic and / or prosodic features. We find that Analor tends to divide speech into smaller segments and that CRF models detect larger segments rather than macro-syntactic periods. However, in general CRF models perform better results than Analor in terms of F-measure.

pdf bib
DNN-based Speech Synthesis Using Abundant Tags of Spontaneous Speech CorpusDNN-based Speech Synthesis Using Abundant Tags of Spontaneous Speech Corpus
Yuki Yamashita | Tomoki Koriyama | Yuki Saito | Shinnosuke Takamichi | Yusuke Ijima | Ryo Masumura | Hiroshi Saruwatari

In this paper, we investigate the effectiveness of using rich annotations in deep neural network (DNN)-based statistical speech synthesis. DNN-based frameworks typically use linguistic information as input features called context instead of directly using text. In such frameworks, we can synthesize not only reading-style speech but also speech with paralinguistic and nonlinguistic features by adding such information to the context. However, it is not clear what kind of information is crucial for reproducing paralinguistic and nonlinguistic features. Therefore, we investigate the effectiveness of rich tags in DNN-based speech synthesis according to the Corpus of Spontaneous Japanese (CSJ), which has a large amount of annotations on paralinguistic features such as prosody, disfluency, and morphological features. Experimental evaluation results shows that the reproducibility of paralinguistic features of synthetic speech was enhanced by adding such information as context.

pdf bib
Lexical Tone Recognition in Mizo using Acoustic-Prosodic Features
Parismita Gogoi | Abhishek Dey | Wendy Lalhminghlui | Priyankoo Sarmah | S R Mahadeva Prasanna

Mizo is an under-studied Tibeto-Burman tonal language of the North-East India. Preliminary research findings have confirmed that four distinct tones of Mizo (High, Low, Rising and Falling) appear in the language. In this work, an attempt is made to automatically recognize four phonological tones in Mizo distinctively using acoustic-prosodic parameters as features. Six features computed from Fundamental Frequency (F0) contours are considered and two classifier models based on Support Vector Machine (SVM) & Deep Neural Network (DNN) are implemented for automatic tonerecognition task respectively. The Mizo database consists of 31950 iterations of the four Mizo tones, collected from 19 speakers using trisyllabic phrases. A four-way classification of tones is attempted with a balanced (equal number of iterations per tone category) dataset for each tone of Mizo. it is observed that the DNN based classifier shows comparable performance in correctly recognizing four phonological Mizo tones as of the SVM based classifier.

pdf bib
Open-source Multi-speaker Speech Corpora for Building Gujarati, Kannada, Malayalam, Marathi, Tamil and Telugu Speech Synthesis SystemsGujarati, Kannada, Malayalam, Marathi, Tamil and Telugu Speech Synthesis Systems
Fei He | Shan-Hui Cathy Chu | Oddur Kjartansson | Clara Rivera | Anna Katanova | Alexander Gutkin | Isin Demirsahin | Cibu Johny | Martin Jansche | Supheakmungkol Sarin | Knot Pipatsrisawat

We present free high quality multi-speaker speech corpora for Gujarati, Kannada, Malayalam, Marathi, Tamil and Telugu, which are six of the twenty two official languages of India spoken by 374 million native speakers. The datasets are primarily intended for use in text-to-speech (TTS) applications, such as constructing multilingual voices or being used for speaker or language adaptation. Most of the corpora (apart from Marathi, which is a female-only database) consist of at least 2,000 recorded lines from female and male native speakers of the language. We present the methodological details behind corpora acquisition, which can be scaled to acquiring data for other languages of interest. We describe the experiments in building a multilingual text-to-speech model that is constructed by combining our corpora. Our results indicate that using these corpora results in good quality voices, with Mean Opinion Scores (MOS) 3.6, for all the languages tested. We believe that these resources, released with an open-source license, and the described methodology will help in the progress of speech applications for the languages described and aid corpora development for other, smaller, languages of India and beyond.

pdf bib
Crowdsourcing Latin American Spanish for Low-Resource Text-to-SpeechLatin American Spanish for Low-Resource Text-to-Speech
Adriana Guevara-Rukoz | Isin Demirsahin | Fei He | Shan-Hui Cathy Chu | Supheakmungkol Sarin | Knot Pipatsrisawat | Alexander Gutkin | Alena Butryna | Oddur Kjartansson

In this paper we present a multidialectal corpus approach for building a text-to-speech voice for a new dialect in a language with existing resources, focusing on various South American dialects of Spanish. We first present public speech datasets for Argentinian, Chilean, Colombian, Peruvian, Puerto Rican and Venezuelan Spanish specifically constructed with text-to-speech applications in mind using crowd-sourcing. We then compare the monodialectal voices built with minimal data to a multidialectal model built by pooling all the resources from all dialects. Our results show that the multidialectal model outperforms the monodialectal baseline models. We also experiment with a zero-resource dialect scenario where we build a multidialectal voice for a dialect while holding out target dialect recordings from the training data.

pdf bib
A Large Scale Speech Sentiment Corpus
Eric Chen | Zhiyun Lu | Hao Xu | Liangliang Cao | Yu Zhang | James Fan

We present a multimodal corpus for sentiment analysis based on the existing Switchboard-1 Telephone Speech Corpus released by the Linguistic Data Consortium. This corpus extends the Switchboard-1 Telephone Speech Corpus by adding sentiment labels from 3 different human annotators for every transcript segment. Each sentiment label can be one of three options : positive, negative, and neutral. Annotators are recruited using Google Cloud’s data labeling service and the labeling task was conducted over the internet. The corpus contains a total of 49500 labeled speech segments covering 140 hours of audio. To the best of our knowledge, this is the largest multimodal Corpus for sentiment analysis that includes both speech and text features.

pdf bib
SibLing Corpus of Russian Dialogue Speech Designed for Research on Speech EntrainmentSibLing Corpus of Russian Dialogue Speech Designed for Research on Speech Entrainment
Tatiana Kachkovskaia | Tatiana Chukaeva | Vera Evdokimova | Pavel Kholiavin | Natalia Kriakina | Daniil Kocharov | Anna Mamushina | Alla Menshikova | Svetlana Zimina

The paper presents a new corpus of dialogue speech designed specifically for research in the field of speech entrainment. Given that the degree of accommodation may depend on a number of social factors, the corpus is designed to encompass 5 types of relations between the interlocutors : those between siblings, close friends, strangers of the same gender, strangers of the other gender, strangers of which one has a higher job position and greater age. Another critical decision taken in this corpus is that in all these social settings one speaker is kept the same. This allows us to trace the changes in his / her speech depending on the interlocutor. The basic set of speakers consists of 10 pairs of same-gender siblings (including 4 pairs of identical twins) aged 23-40, and each of them was recorded in the 5 settings mentioned above. In total we obtained 90 dialogues of 25-60 minutes each. The speakers played a card game and a map game ; they were recorded in a soundproof studio without being able to see each other due to a non-transparent screen between them. The corpus contains orthographic, phonetic and prosodic annotation and is segmented into turns and inter-pausal units.

pdf bib
PhonBank and Data Sharing : Recent Developments in European PortuguesePhonBank and Data Sharing: Recent Developments in European Portuguese
Ana Margarida Ramalho | Maria João Freitas | Yvan Rose

84 104 105 115 32 112 97 112 101 114 32 112 114 101 115 101 110 116 115 32 116 104 101 32 114 101 99 101 110 116 108 121 32 112 117 98 108 105 115 104 101 100 32 82 65 77 65 76 72 79 45 69 80 32 97 110 100 32 80 72 79 78 79 68 73 83 32 99 111 114 112 111 114 97 46 32 66 111 116 104 32 105 110 99 108 117 100 101 32 69 117 114 111 112 101 97 110 32 80 111 114 116 117 103 117 101 115 101 32 112 114 111 100 117 99 116 105 111 110 32 100 97 116 97 32 102 114 111 109 32 80 111 114 116 117 103 117 101 115 101 32 99 104 105 108 100 114 101 110 32 119 105 116 104 32 116 121 112 105 99 97 108 32 40 82 65 77 65 76 72 79 45 69 80 41 32 97 110 100 32 112 114 111 116 114 97 99 116 101 100 32 40 80 72 79 78 79 68 73 83 41 32 112 104 111 110 111 108 111 103 105 99 97 108 32 100 101 118 101 108 111 112 109 101 110 116 46 32 84 104 101 32 100 97 116 97 32 105 110 32 116 104 101 32 116 119 111 32 99 111 114 112 111 114 97 32 119 101 114 101 32 99 111 108 108 101 99 116 101 100 32 117 115 105 110 103 32 116 104 101 32 112 104 111 110 111 108 111 103 105 99 97 108 32 97 115 115 101 115 115 109 101 110 116 32 116 111 111 108 32 67 76 67 80 45 69 80 44 32 100 101 118 101 108 111 112 101 100 32 105 110 32 116 104 101 32 99 111 110 116 101 120 116 32 111 102 32 116 104 101 32 67 114 111 115 115 108 105 110 103 117 105 115 116 105 99 32 67 104 105 108 100 32 80 104 111 110 111 108 111 103 121 32 80 114 111 106 101 99 116 44 32 99 111 111 114 100 105 110 97 116 101 100 32 98 121 32 66 97 114 98 97 114 97 32 66 101 114 110 104 97 114 100 116 32 97 110 100 32 74 111 101 32 83 116 101 109 98 101 114 103 101 114 32 40 85 110 105 118 101 114 115 105 116 121 32 111 102 32 66 114 105 116 105 115 104 32 67 111 108 117 109 98 105 97 32 40 85 66 67 41 44 32 67 97 110 97 100 97 41 46 32 32 66 111 116 104 32 99 111 114 112 111 114 97 32 97 114 101 32 112 97 114 116 32 111 102 32 116 104 101 32 80 104 111 110 66 97 110 107 32 80 114 111 106 101 99 116 32 40 66 114 105 97 110 32 77 97 99 87 104 105 110 110 101 121 32 40 67 97 114 110 101 103 105 101 32 77 101 108 108 111 110 44 32 85 83 65 41 32 97 110 100 32 89 118 97 110 32 82 111 115 101 32 40 77 101 109 111 114 105 97 108 32 85 110 105 118 101 114 115 105 116 121 32 111 102 32 78 101 119 102 111 117 110 100 108 97 110 100 44 32 67 97 110 97 100 97 41 44 32 119 104 105 99 104 32 105 115 32 116 104 101 32 99 104 105 108 100 32 112 104 111 110 111 108 111 103 121 32 99 111 109 112 111 110 101 110 116 32 111 102 32 84 97 108 107 66 97 110 107 44 32 99 111 111 114 100 105 110 97 116 101 100 32 98 121 32 66 114 105 97 110 32 77 97 99 87 104 105 110 110 101 121 46 32 84 104 101 32 100 97 116 97 32 97 116 32 80 104 111 110 66 97 110 107 32 105 115 32 101 100 105 116 101 100 32 105 110 32 80 104 111 110 32 102 111 114 109 97 116 44 32 97 32 108 97 110 103 117 97 103 101 32 116 111 111 108 32 100 101 115 105 103 110 101 100 32 97 110 100 32 98 117 105 108 116 32 98 121 32 89 118 97 110 32 82 111 115 101 32 97 110 100 32 71 114 101 103 32 72 101 100 108 117 110 100 32 40 77 101 109 111 114 105 97 108 32 85 110 105 118 101 114 115 105 116 121 32 111 102 32 78 101 119 102 111 117 110 100 108 97 110 100 41 32 97 110 100 32 119 105 100 101 108 121 32 117 115 101 100 32 98 121 32 114 101 115 101 97 114 99 104 101 114 115 32 119 111 114 107 105 110 103 32 105 110 32 116 104 101 32 102 105 101 108 100 32 111 102 32 112 104 111 110 111 108 111 103 105 99 97 108 32 97 99 113 117 105 115 105 116 105 111 110 46 32 82 65 77 65 76 72 79 45 69 80 32 99 111 110 116 97 105 110 115 32 112 114 111 100 117 99 116 105 111 110 32 100 97 116 97 32 102 114 111 109 32 56 55 32 116 121 112 105 99 97 108 108 121 32 100 101 118 101 108 111 112 105 110 103 32 99 104 105 108 100 114 101 110 44 32 97 103 101 100 32 50 59 49 49 32 116 111 32 54 59 48 52 44 32 97 108 108 32 109 111 110 111 108 105 110 103 117 97 108 115 46 32 80 72 79 78 79 68 73 83 32 105 110 99 108 117 100 101 115 32 112 114 111 100 117 99 116 105 111 110 32 100 97 116 97 32 102 114 111 109 32 50 50 32 99 104 105 108 100 114 101 110 32 100 105 97 103 110 111 115 101 100 32 119 105 116 104 32 100 105 102 102 101 114 101 110 116 32 116 121 112 101 115 32 111 102 32 115 112 101 101 99 104 32 97 110 100 32 108 97 110 103 117 97 103 101 32 100 105 115 111 114 100 101 114 115 44 32 97 108 108 32 69 80 32 109 111 110 111 108 105 110 103 117 97 108 115 44 32 97 103 101 100 32 51 59 50 32 116 111 32 49 49 44 48 53 46 32 66 111 116 104 32 99 111 114 112 111 114 97 32 97 114 101 32 111 112 101 110 32 97 99 99 101 115 115 32 108 97 110 103 117 97 103 101 32 114 101 115 111 117 114 99 101 115 32 97 110 100 32 99 111 110 116 114 105 98 117 116 101 32 116 111 32 101 110 108 97 114 103 101 32 116 104 101 32 97 109 111 117 110 116 32 111 102 32 112 114 111 100 117 99 116 105 111 110 32 100 97 116 97 32 111 110 32 116 104 101 32 97 99 113 117 105 115 105 116 105 111 110 32 111 102 32 69 117 114 111 112 101 97 110 32 80 111 114 116 117 103 117 101 115 101 32 97 118 97 105 108 97 98 108 101 32 105 110 32 80 104 111 110 66 97 110 107 46

pdf bib
SMASH Corpus : A Spontaneous Speech Corpus Recording Third-person Audio Commentaries on GameplaySMASH Corpus: A Spontaneous Speech Corpus Recording Third-person Audio Commentaries on Gameplay
Yuki Saito | Shinnosuke Takamichi | Hiroshi Saruwatari

Developing a spontaneous speech corpus would be beneficial for spoken language processing and understanding. We present a speech corpus named the SMASH corpus, which includes spontaneous speech of two Japanese male commentators that made third-person audio commentaries during the gameplay of a fighting game. Each commentator ad-libbed while watching the gameplay with various topics covering not only explanations of each moment to convey the information on the fight but also comments to entertain listeners. We made transcriptions and topic tags as annotations on the recorded commentaries with our two-step method. We first made automatic and manual transcriptions of the commentaries and then manually annotated the topic tags. This paper describes how we constructed the SMASH corpus and reports some results of the annotations.

pdf bib
Gender Representation in Open Source Speech Resources
Mahault Garnerin | Solange Rossato | Laurent Besacier

With the rise of artificial intelligence (AI) and the growing use of deep-learning architectures, the question of ethics, transparency and fairness of AI systems has become a central concern within the research community. We address transparency and fairness in spoken language systems by proposing a study about gender representation in speech resources available through the Open Speech and Language Resource platform. We show that finding gender information in open source corpora is not straightforward and that gender balance depends on other corpus characteristics (elicited / non elicited speech, low / high resource language, speech task targeted). The paper ends with recommendations about metadata and gender information for researchers in order to assure better transparency of the speech systems built using such corpora.

pdf bib
Call My Net 2 : A New Resource for Speaker Recognition
Karen Jones | Stephanie Strassel | Kevin Walker | Jonathan Wright

We introduce the Call My Net 2 (CMN2) Corpus, a new resource for speaker recognition featuring Tunisian Arabic conversations between friends and family, incorporating both traditional telephony and VoIP data. The corpus contains data from over 400 Tunisian Arabic speakers collected via a custom-built platform deployed in Tunis, with each speaker making 10 or more calls each lasting up to 10 minutes. Calls include speech in various realistic and natural acoustic settings, both noisy and non-noisy. Speakers used a variety of handsets, including landline and mobile devices, and made VoIP calls from tablets or computers. All calls were subject to a series of manual and automatic quality checks, including speech duration, audio quality, language identity and speaker identity. The CMN2 corpus has been used in two NIST Speaker Recognition Evaluations (SRE18 and SRE19), and the SRE test sets as well as the full CMN2 corpus will be published in the Linguistic Data Consortium Catalog. We describe CMN2 corpus requirements, the telephone collection platform, and procedures for call collection. We review properties of the CMN2 dataset and discuss features of the corpus that distinguish it from prior SRE collection efforts, including some of the technical challenges encountered with collecting VoIP data.

pdf bib
Development and Evaluation of Speech Synthesis Corpora for LatvianLatvian
Roberts Darģis | Peteris Paikens | Normunds Gruzitis | Ilze Auzina | Agate Akmane

Text to speech (TTS) systems are necessary for all languages to ensure accessibility and availability of digital language services. Recent advances in neural speech synthesis have eText to speech (TTS) systems are necessary for any language to ensure accessibility and availability of digital language services. Recent advances in neural speech synthesis have enabled the development of such systems with a data-driven approach that does not require significant development of language-specific tools. However, smaller languages often lack speech corpora that would be sufficient for training current neural TTS models, which require at least 30 hours of good quality audio recordings from a single speaker in a noiseless environment with matching transcriptions. Making such a corpus manually can be cost prohibitive. This paper presents an unsupervised approach to obtain a suitable corpus from unannotated recordings using automated speech recognition for transcription, as well as automated speaker segmentation and identification. The proposed method and software tools are applied and evaluated on a case study for developing a corpus suitable for Latvian speech synthesis based on Latvian public radio archive data.nabled the development of such systems with a data-driven approach that does not require much language-specific tool development. However, smaller languages often lack speech corpora that would be sufficient for training current neural TTS models, which require approximately 30 hours of good quality audio recordings from a single speaker in a noiseless environment with matching transcriptions. Making such a corpus manually can be cost prohibitive. This paper presents an unsupervised approach to obtain a suitable corpus from unannotated recordings using automated speech recognition for transcription, as well as automated speaker segmentation and identification. The proposed methods and software tools are applied and evaluated on a case study for developing a corpus suitable for Latvian speech synthesis based on Latvian public radio archive data.

pdf bib
GameWikiSum : a Novel Large Multi-Document Summarization DatasetGameWikiSum: a Novel Large Multi-Document Summarization Dataset
Diego Antognini | Boi Faltings

Today’s research progress in the field of multi-document summarization is obstructed by the small number of available datasets. Since the acquisition of reference summaries is costly, existing datasets contain only hundreds of samples at most, resulting in heavy reliance on hand-crafted features or necessitating additional, manually annotated data. The lack of large corpora therefore hinders the development of sophisticated models. Additionally, most publicly available multi-document summarization corpora are in the news domain, and no analogous dataset exists in the video game domain. In this paper, we propose GameWikiSum, a new domain-specific dataset for multi-document summarization, which is one hundred times larger than commonly used datasets, and in another domain than news. Input documents consist of long professional video game reviews as well as references of their gameplay sections in Wikipedia pages. We analyze the proposed dataset and show that both abstractive and extractive models can be trained on it. We release GameWikiSum for further research :

pdf bib
Summarization Beyond News : The Automatically Acquired Fandom Corpora
Benjamin Hättasch | Nadja Geisler | Christian M. Meyer | Carsten Binnig

Large state-of-the-art corpora for training neural networks to create abstractive summaries are mostly limited to the news genre, as it is expensive to acquire human-written summaries for other types of text at a large scale. In this paper, we present a novel automatic corpus construction approach to tackle this issue as well as three new large open-licensed summarization corpora based on our approach that can be used for training abstractive summarization models. Our constructed corpora contain fictional narratives, descriptive texts, and summaries about movies, television, and book series from different domains. All sources use a creative commons (CC) license, hence we can provide the corpora for download. In addition, we also provide a ready-to-use framework that implements our automatic construction approach to create custom corpora with desired parameters like the length of the target summary and the number of source documents from which to create the summary. The main idea behind our automatic construction approach is to use existing large text collections (e.g., thematic wikis) and automatically classify whether the texts can be used as (query-focused) multi-document summaries and align them with potential source texts. As a final contribution, we show the usefulness of our automatic construction approach by running state-of-the-art summarizers on the corpora and through a manual evaluation with human annotators.

pdf bib
Invisible to People but not to Machines : Evaluation of Style-aware HeadlineGeneration in Absence of Reliable Human JudgmentHeadlineGeneration in Absence of Reliable Human Judgment
Lorenzo De Mattei | Michele Cafagna | Felice Dell’Orletta | Malvina Nissim

We automatically generate headlines that are expected to comply with the specific styles of two different Italian newspapers. Through a data alignment strategy and different training / testing settings, we aim at decoupling content from style and preserve the latter in generation. In order to evaluate the generated headlines’ quality in terms of their specific newspaper-compliance, we devise a fine-grained evaluation strategy based on automatic classification. We observe that our models do indeed learn newspaper-specific style. Importantly, we also observe that humans are n’t reliable judges for this task, since although familiar with the newspapers, they are not able to discern their specific styles even in the original human-written headlines. The utility of automatic evaluation goes therefore beyond saving the costs and hurdles of manual annotation, and deserves particular care in its design.

pdf bib
Align then Summarize : Automatic Alignment Methods for Summarization Corpus Creation
Paul Tardy | David Janiszek | Yannick Estève | Vincent Nguyen

Summarizing texts is not a straightforward task. Before even considering text summarization, one should determine what kind of summary is expected. How much should the information be compressed? Is it relevant to reformulate or should the summary stick to the original phrasing? State-of-the-art on automatic text summarization mostly revolves around news articles. We suggest that considering a wider variety of tasks would lead to an improvement in the field, in terms of generalization and robustness. We explore meeting summarization : generating reports from automatic transcriptions. Our work consists in segmenting and aligning transcriptions with respect to reports, to get a suitable dataset for neural summarization. Using a bootstrapping approach, we provide pre-alignments that are corrected by human annotators, making a validation set against which we evaluate automatic models. This consistently reduces annotators’ efforts by providing iteratively better pre-alignment and maximizes the corpus size by using annotations from our automatic alignment models. Evaluation is conducted on publicmeetings, a novel corpus of aligned public meetings. We report automatic alignment and summarization performances on this corpus and show that automatic alignment is relevant for data annotation since it leads to large improvement of almost +4 on all ROUGE scores on the summarization task.

pdf bib
Diverging Divergences : Examining Variants of Jensen Shannon Divergence for Corpus Comparison TasksJensen Shannon Divergence for Corpus Comparison Tasks
Jinghui Lu | Maeve Henchion | Brian Mac Namee

Jensen-Shannon divergence (JSD) is a distribution similarity measurement widely used in natural language processing. In corpus comparison tasks, where keywords are extracted to reveal the divergence between different corpora (for example, social media posts from proponents of different views on a political issue), two variants of JSD have emerged in the literature. One of these uses a weighting based on the relative sizes of the corpora being compared. In this paper we argue that this weighting is unnecessary and, in fact, can lead to misleading results. We recommend that this weighted version is not used. We base this recommendation on an analysis of the JSD variants and experiments showing how they impact corpus comparison results as the relative sizes of the corpora being compared change.

pdf bib
SC-CoMIcs : A Superconductivity Corpus for Materials InformaticsSC-CoMIcs: A Superconductivity Corpus for Materials Informatics
Kyosuke Yamaguchi | Ryoji Asahi | Yutaka Sasaki

This paper describes a novel corpus tailored for the text mining of superconducting materials in Materials Informatics (MI), named SuperConductivety Corpus for Materials Informatics (SC-CoMIcs). Different from biomedical informatics, there exist very few corpora targeting Materials Science and Engineering (MSE). Especially, there is no sizable corpus which can be used to assist the search of superconducting materials. A team of materials scientists and natural language processing experts jointly designed the annotation and constructed a corpus consisting of manually-annotated 1,000 MSE abstracts related to superconductivity. We conducted experiments on the corpus with a neural Named Entity Recognition (NER) tool. The experimental results show that NER performance over the corpus is around 77 % in terms of micro-F1, which is comparable to human annotator agreement rates. Using the trained NER model, we automatically annotated 9,000 abstracts and created a term retrieval tool based on the term similarity. This tool can find superconductivity terms relevant to a query term within a specified Named Entity category, which demonstrates the power of our SC-CoMIcs, efficiently providing knowledge for Materials Informatics applications from rapidly expanding publications.

pdf bib
Japanese Realistic Textual Entailment CorpusJapanese Realistic Textual Entailment Corpus
Yuta Hayashibe

We perform the textual entailment (TE) corpus construction for the Japanese Language with the following three characteristics : First, the corpus consists of realistic sentences ; that is, all sentences are spontaneous or almost equivalent. It does not need manual writing which causes hidden biases. Second, the corpus contains adversarial examples. We collect challenging examples that can not be solved by a recent pre-trained language model. Third, the corpus contains explanations for a part of non-entailment labels. We perform the reasoning annotation where annotators are asked to check which tokens in hypotheses are the reason why the relations are labeled. It makes easy to validate the annotation and analyze system errors. The resulting corpus consists of 48,000 realistic Japanese examples. It is the largest among publicly available Japanese TE corpora. Additionally, it is the first Japanese TE corpus that includes reasons for the annotation as we know. We are planning to distribute this corpus to the NLP community at the time of publication.

pdf bib
HypoNLI : Exploring the Artificial Patterns of Hypothesis-only Bias in Natural Language InferenceHypoNLI: Exploring the Artificial Patterns of Hypothesis-only Bias in Natural Language Inference
Tianyu Liu | Zheng Xin | Baobao Chang | Zhifang Sui

Many recent studies have shown that for models trained on datasets for natural language inference (NLI), it is possible to make correct predictions by merely looking at the hypothesis while completely ignoring the premise. In this work, we manage to derive adversarial examples in terms of the hypothesis-only bias and explore eligible ways to mitigate such bias. Specifically, we extract various phrases from the hypotheses (artificial patterns) in the training sets, and show that they have been strong indicators to the specific labels. We then figure out ‘hard’ and ‘easy’ instances from the original test sets whose labels are opposite to or consistent with those indications. We also set up baselines including both pretrained models (BERT, RoBerta, XLNet) and competitive non-pretrained models (InferSent, DAM, ESIM). Apart from the benchmark and baselines, we also investigate two debiasing approaches which exploit the artificial pattern modeling to mitigate such hypothesis-only bias : down-sampling and adversarial training. We believe those methods can be treated as competitive baselines in NLI debiasing tasks.

pdf bib
TaPaCo : A Corpus of Sentential Paraphrases for 73 LanguagesTaPaCo: A Corpus of Sentential Paraphrases for 73 Languages
Yves Scherrer

This paper presents TaPaCo, a freely available paraphrase corpus for 73 languages extracted from the Tatoeba database. Tatoeba is a crowdsourcing project mainly geared towards language learners. Its aim is to provide example sentences and translations for particular linguistic constructions and words. The paraphrase corpus is created by populating a graph with Tatoeba sentences and equivalence links between sentences meaning the same thing. This graph is then traversed to extract sets of paraphrases. Several language-independent filters and pruning steps are applied to remove uninteresting sentences. A manual evaluation performed on three languages shows that between half and three quarters of inferred paraphrases are correct and that most remaining ones are either correct but trivial, or near-paraphrases that neutralize a morphological distinction. The corpus contains a total of 1.9 million sentences, with 200-250 000 sentences per language. It covers a range of languages for which, to our knowledge, no other paraphrase dataset exists. The dataset is available at

pdf bib
Towards the Necessity for Debiasing Natural Language Inference Datasets
Mithun Paul Panenghat | Sandeep Suntwal | Faiz Rafique | Rebecca Sharp | Mihai Surdeanu

Modeling natural language inference is a challenging task. With large annotated data sets available it has now become feasible to train complex neural network based inference methods which achieve state of the art performance. However, it has been shown that these models also learn from the subtle biases inherent in these datasets (CITATION). In this work we explore two techniques for delexicalization that modify the datasets in such a way that we can control the importance that neural-network based methods place on lexical entities. We demonstrate that the proposed methods not only maintain the performance in-domain but also improve performance in some out-of-domain settings. For example, when using the delexicalized version of the FEVER dataset, the in-domain performance of a state of the art neural network method dropped only by 1.12 % while its out-of-domain performance on the FNC dataset improved by 4.63 %. We release the delexicalized versions of three common datasets used in natural language inference. These datasets are delexicalized using two methods : one which replaces the lexical entities in an overlap-aware manner, and a second, which additionally incorporates semantic lifting of nouns and verbs to their WordNet hypernym synsets

pdf bib
A French Corpus for Semantic SimilarityFrench Corpus for Semantic Similarity
Rémi Cardon | Natalia Grabar

Semantic similarity is an area of Natural Language Processing that is useful for several downstream applications, such as machine translation, natural language generation, information retrieval, or question answering. The task consists in assessing the extent to which two sentences express or do not express the same meaning. To do so, corpora with graded pairs of sentences are required. The grade is positioned on a given scale, usually going from 0 (completely unrelated) to 5 (equivalent semantics). In this work, we introduce such a corpus for French, the first that we know of. It is comprised of 1,010 sentence pairs with grades from five annotators. We describe the annotation process, analyse these data, and perform a few experiments for the automatic grading of semantic similarity.

pdf bib
TIARA : A Tool for Annotating Discourse Relations and Sentence ReorderingTIARA: A Tool for Annotating Discourse Relations and Sentence Reordering
Jan Wira Gotama Putra | Simone Teufel | Kana Matsumura | Takenobu Tokunaga

This paper introduces TIARA, a new publicly available web-based annotation tool for discourse relations and sentence reordering. Annotation tasks such as these, which are based on relations between large textual objects, are inherently hard to visualise without either cluttering the display and/or confusing the annotators. TIARA deals with the visual complexity during the annotation process by systematically simplifying the layout, and by offering interactive visualisation, including coloured links, indentation, and dual-view. TIARA’s text view allows annotators to focus on the analysis of logical sequencing between sentences. A separate tree view allows them to review their analysis in terms of the overall discourse structure. The dual-view gives it an edge over other discourse annotation tools and makes it particularly attractive as an educational tool (e.g., for teaching students how to argue more effectively). As it is based on standard web technologies and can be easily customised to other annotation schemes, it can be easily used by anybody. Apart from the project it was originally designed for, in which hundreds of texts were annotated by three annotators, TIARA has already been adopted by a second discourse annotation study, which uses it in the teaching of argumentation.

pdf bib
ThaiLMCut : Unsupervised Pretraining for Thai Word SegmentationThaiLMCut: Unsupervised Pretraining for Thai Word Segmentation
Suteera Seeha | Ivan Bilan | Liliana Mamani Sanchez | Johannes Huber | Michael Matuschek | Hinrich Schütze

We propose ThaiLMCut, a semi-supervised approach for Thai word segmentation which utilizes a bi-directional character language model (LM) as a way to leverage useful linguistic knowledge from unlabeled data. After the language model is trained on substantial unlabeled corpora, the weights of its embedding and recurrent layers are transferred to a supervised word segmentation model which continues fine-tuning them on a word segmentation task. Our experimental results demonstrate that applying the LM always leads to a performance gain, especially when the amount of labeled data is small. In such cases, the F1 Score increased by up to 2.02 %. Even on abig labeled dataset, a small improvement gain can still be obtained. The approach has also shown to be very beneficial for out-of-domain settings with a gain in F1 Score of up to 3.13 %. Finally, we show that ThaiLMCut can outperform other open source state-of-the-art models achieving an F1 Score of 98.78 % on the standard benchmark, InterBEST2009.

pdf bib
CCOHA : Clean Corpus of Historical American EnglishCCOHA: Clean Corpus of Historical American English
Reem Alatrash | Dominik Schlechtweg | Jonas Kuhn | Sabine Schulte im Walde

Modelling language change is an increasingly important area of interest within the fields of sociolinguistics and historical linguistics. In recent years, there has been a growing number of publications whose main concern is studying changes that have occurred within the past centuries. The Corpus of Historical American English (COHA) is one of the most commonly used large corpora in diachronic studies in English. This paper describes methods applied to the downloadable version of the COHA corpus in order to overcome its main limitations, such as inconsistent lemmas and malformed tokens, without compromising its qualitative and distributional properties. The resulting corpus CCOHA contains a larger number of cleaned word tokens which can offer better insights into language change and allow for a larger variety of tasks to be performed.

pdf bib
Outbound Translation User Interface Ptakopt : A Pilot Study
Vilém Zouhar | Ondřej Bojar

It is not uncommon for Internet users to have to produce a text in a foreign language they have very little knowledge of and are unable to verify the translation quality. We call the task outbound translation and explore it by introducing an open-source modular system Ptakopt. Its main purpose is to inspect human interaction with MT systems enhanced with additional subsystems, such as backward translation and quality estimation. We follow up with an experiment on (Czech) human annotators tasked to produce questions in a language they do not speak (German), with the help of Ptakopt. We focus on three real-world use cases (communication with IT support, describing administrative issues and asking encyclopedic questions) from which we gain insight into different strategies users take when faced with outbound translation tasks. Round trip translation is known to be unreliable for evaluating MT systems but our experimental evaluation documents that it works very well for users, at least on MT systems of mid-range quality.

pdf bib
Seshat : a Tool for Managing and Verifying Annotation Campaigns of Audio DataSeshat: a Tool for Managing and Verifying Annotation Campaigns of Audio Data
Hadrien Titeux | Rachid Riad | Xuan-Nga Cao | Nicolas Hamilakis | Kris Madden | Alejandrina Cristia | Anne-Catherine Bachoud-Lévi | Emmanuel Dupoux

We introduce Seshat, a new, simple and open-source software to efficiently manage annotations of speech corpora. The Seshat software allows users to easily customise and manage annotations of large audio corpora while ensuring compliance with the formatting and naming conventions of the annotated output files. In addition, it includes procedures for checking the content of annotations following specific rules that can be implemented in personalised parsers. Finally, we propose a double-annotation mode, for which Seshat computes automatically an associated inter-annotator agreement with the gamma measure taking into account the categorisation and segmentation discrepancies.

pdf bib
KonText : Advanced and Flexible Corpus Query InterfaceKonText: Advanced and Flexible Corpus Query Interface
Tomáš Machálek

We present an advanced, highly customizable corpus query interface KonText built on top of core libraries of the open-source corpus search engine NoSketch Engine (NoSkE). The aim is to overcome some limitations of the original NoSkE user interface and provide integration capabilities allowing connection of the basic search service with other language resources (LRs). The introduced features are based on long-term feedback given by the users and researchers of the Czech National Corpus (CNC) along with other LRs providers running KonText as a part of their services. KonText is a fully operational and mature software deployed at the CNC since 2014 that currently handles thousands user queries per day.

pdf bib
RKorAPClient : An R Package for Accessing the German Reference Corpus DeReKo via KorAPRKorAPClient: An R Package for Accessing the German Reference Corpus DeReKo via KorAP
Marc Kupietz | Nils Diewald | Eliza Margaretha

Making corpora accessible and usable for linguistic research is a huge challenge in view of (too) big data, legal issues and a rapidly evolving methodology. This does not only affect the design of user-friendly graphical interfaces to corpus analysis tools, but also the availability of programming interfaces supporting access to the functionality of these tools from various analysis and development environments. RKorAPClient is a new research tool in the form of an R package that interacts with the Web API of the corpus analysis platform KorAP, which provides access to large annotated corpora, including the German reference corpus DeReKo with 45 billion tokens. In addition to optionally authenticated KorAP API access, RKorAPClient provides further processing and visualization features to simplify common corpus analysis tasks. This paper introduces the basic functionality of RKorAPClient and exemplifies various analysis tasks based on DeReKo, that are bundled within the R package and can serve as a basic framework for advanced analysis and visualization approaches.

pdf bib
The xtsv Framework and the Twelve Virtues of Pipelines
Balázs Indig | Bálint Sass | Iván Mittelholcz

We present xtsv, an abstract framework for building NLP pipelines. It covers several kinds of functionalities which can be implemented at an abstract level. We survey these features and argue that all are desired in a modern pipeline. The framework has a simple yet powerful internal communication format which is essentially tsv (tab separated values) with header plus some additional features. We put emphasis on the capabilities of the presented framework, for example its ability to allow new modules to be easily integrated or replaced, or the variety of its usage options. When a module is put into xtsv, all functionalities of the system are immediately available for that module, and the module can be be a part of an xtsv pipeline. The design also allows convenient investigation and manual correction of the data flow from one module to another. We demonstrate the power of our framework with a successful application : a concrete NLP pipeline for Hungarian called e-magyar text processing system (emtsv) which integrates Hungarian NLP tools in xtsv. All the advantages of the pipeline come from the inherent properties of the xtsv framework.

pdf bib
Data Query Language and Corpus Tools for Slot-Filling and Intent Classification Data
Stefan Larson | Eric Guldan | Kevin Leach

Typical machine learning approaches to developing task-oriented dialog systems require the collection and management of large amounts of training data, especially for the tasks of intent classification and slot-filling. Managing this data can be cumbersome without dedicated tools to help the dialog system designer understand the nature of the data. This paper presents a toolkit for analyzing slot-filling and intent classification corpora. We present a toolkit that includes (1) a new lightweight and readable data and file format for intent classification and slot-filling corpora, (2) a new query language for searching intent classification and slot-filling corpora, and (3) tools for understanding the structure and makeup for such corpora. We apply our toolkit to several well-known NLU datasets, and demonstrate that our toolkit can be used to uncover interesting and surprising insights. By releasing our toolkit to the research community, we hope to enable others to develop more robust and intelligent slot-filling and intent classification models.

pdf bib
SHR++ : An Interface for Morpho-syntactic Annotation of Sanskrit CorporaSHR++: An Interface for Morpho-syntactic Annotation of Sanskrit Corpora
Amrith Krishna | Shiv Vidhyut | Dilpreet Chawla | Sruti Sambhavi | Pawan Goyal

We propose a web-based annotation framework, SHR++, for morpho-syntactic annotation of corpora in Sanskrit. SHR++ is designed to generate annotations for the word-segmentation, morphological parsing and dependency analysis tasks in Sanskrit. It incorporates analyses and predictions from various tools designed for processing texts in Sanskrit, and utilise them to ease the cognitive load of the human annotators. Specifically, SHR++ uses Sanskrit Heritage Reader, a lexicon driven shallow parser for enumerating all the phonetically and lexically valid word splits along with their morphological analyses for a given string. This would help the annotators in choosing the solutions, rather than performing the segmentations by themselves. Further, predictions from a word segmentation tool are added as suggestions that can aid the human annotators in their decision making. Our evaluation shows that enabling this segmentation suggestion component reduces the annotation time by 20.15 %. SHR++ can be accessed online at and the codebase, for the independent deployment of the system elsewhere, is hosted at

pdf bib
Gamification Platform for Collecting Task-oriented Dialogue Data
Haruna Ogawa | Hitoshi Nishikawa | Takenobu Tokunaga | Hikaru Yokono

Demand for massive language resources is increasing as the data-driven approach has established a leading position in Natural Language Processing. However, creating dialogue corpora is still a difficult task due to the complexity of the human dialogue structure and the diversity of dialogue topics. Though crowdsourcing is majorly used to assemble such data, it presents problems such as less-motivated workers. We propose a platform for collecting task-oriented situated dialogue data by using gamification. Combining a video game with data collection benefits such as motivating workers and cost reduction. Our platform enables data collectors to create their original video game in which they can collect dialogue data of various types of tasks by using the logging function of the platform. Also, the platform provides the annotation function that enables players to annotate their own utterances. The annotation can be gamified aswell. We aim at high-quality annotation by introducing such self-annotation method. We implemented a prototype of the proposed platform and conducted a preliminary evaluation to obtain promising results in terms of both dialogue data collection and self-annotation.

pdf bib
Improving Sentence Boundary Detection for Spoken Language Transcripts
Ines Rehbein | Josef Ruppenhofer | Thomas Schmidt

This paper presents experiments on sentence boundary detection in transcripts of spoken dialogues. Segmenting spoken language into sentence-like units is a challenging task, due to disfluencies, ungrammatical or fragmented structures and the lack of punctuation. In addition, one of the main bottlenecks for many NLP applications for spoken language is the small size of the training data, as the transcription and annotation of spoken language is by far more time-consuming and labour-intensive than processing written language. We therefore investigate the benefits of data expansion and transfer learning and test different ML architectures for this task. Our results show that data expansion is not straightforward and even data from the same domain does not always improve results. They also highlight the importance of modelling, i.e. of finding the best architecture and data representation for the task at hand. For the detection of boundaries in spoken language transcripts, we achieve a substantial improvement when framing the boundary detection problem assentence pair classification task, as compared to a sequence tagging approach.

pdf bib
PyVallex : A Processing System for Valency Lexicon DataPyVallex: A Processing System for Valency Lexicon Data
Jonathan Verner | Anna Vernerová

PyVallex is a Python-based system for presenting, searching / filtering, editing / extending and automatic processing of machine-readable lexicon data originally available in a text-based format. The system consists of several components : a parser for the specific lexicon format used in several valency lexicons, a data-validation framework, a regular expression based search engine, a map-reduce style framework for querying the lexicon data and a web-based interface integrating complex search and some basic editing capabilities. PyVallex provides most of the typical functionalities of a Dictionary Writing System (DWS), such as multiple presentation modes for the underlying lexical database, automatic evaluation of consistency tests, and a mechanism of merging updates coming from multiple sources. The editing functionality is currently limited to the client-side interface and edits of existing lexical entries, but additional script-based operations on the database are also possible. The code is published under the open source MIT license and is also available in the form of a Python module for integrating into other software.

pdf bib
Fintan-Flexible, Integrated Transformation and Annotation eNgineeringNgineering
Christian Fäth | Christian Chiarcos | Björn Ebbrecht | Maxim Ionov

We introduce the Flexible and Integrated Transformation and Annotation eNgeneering (Fintan) platform for converting heterogeneous linguistic resources to RDF. With its modular architecture, workflow management and visualization features, Fintan facilitates the development of complex transformation pipelines by integrating generic RDF converters and augmenting them with extended graph processing capabilities : Existing converters can be easily deployed to the system by means of an ontological data structure which renders their properties and the dependencies between transformation steps. Development of subsequent graph transformation steps for resource transformation, annotation engineering or entity linking is further facilitated by a novel visual rendering of SPARQL queries. A graphical workflow manager allows to easily manage the converter modules and combine them to new transformation pipelines. Employing the stream-based graph processing approach first implemented with CoNLL-RDF, we address common challenges and scalability issues when transforming resources and showcase the performance of Fintan by means of a purely graph-based transformation of the Universal Morphology data to RDF.

pdf bib
Interchange Formats for Visualization : LIF and MMIFLIF and MMIF
Kyeongmin Rim | Kelley Lynch | Marc Verhagen | Nancy Ide | James Pustejovsky

Promoting interoperrable computational linguistics (CL) and natural language processing (NLP) application platforms and interchange-able data formats have contributed improving discoverabilty and accessbility of the openly available NLP software. In this paper, wediscuss the enhanced data visualization capabilities that are also enabled by inter-operating NLP pipelines and interchange formats. For adding openly available visualization tools and graphical annotation tools to the Language Applications Grid (LAPPS Grid) andComputational Linguistics Applications for Multimedia Services (CLAMS) toolboxes, we have developed interchange formats that cancarry annotations and metadata for text and audiovisual source data. We descibe those data formats and present case studies where wesuccessfully adopt open-source visualization tools and combine them with CL tools.