LREC 2020 Paper Dissemination (7/10)

LREC 2020 was not held in Marseille this year and only the Proceedings were published.

The ELRA Board and the LREC 2020 Programme Committee now feel that those papers should be disseminated again, in a thematic-oriented way, shedding light on specific “topics/sessions”.

Packages with several sessions will be disseminated every Tuesday for 10 weeks, from Nov 10, 2020 until the end of January 2021.

Each session displays papers’ title and authors, with corresponding abstract (for ease of reading) and url, in like manner as the Book of Abstracts we used to print and distribute at LRECs.

We hope that you discover interesting, even exciting, work that may be useful for your own research.

Group of papers sent on January 5, 2021

Links to each session

MultiWord Expressions & Collocations

A Domain-Specific Dataset of Difficulty Ratings for German Noun Compounds in the Domains DIY, Cooking and Automotive

Julia Bettinger, Anna Hätty, Michael Dorna and 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.


All That Glitters is Not Gold: A Gold Standard of Adjective-Noun Collocations for German

Yana Strakatova, Neele Falk, Isabel Fuhrmann, Erhard Hinrichs and 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.


Variants of Vector Space Reductions for Predicting the Compositionality of English Noun Compounds

Pegah Alipoor and Sabine Schulte im Walde

Predicting the degree of compositionality of noun compounds such as “snowball” and “butterfly” is a crucial ingredient for lexicography and Natural Language Processing applications, to know whether the compound should be treated as a whole, or through its constituents, and what it means. Computational approaches for an automatic prediction typically represent and compare compounds and their constituents within a vector space and use distributional similarity as a proxy to predict the semantic relatedness between the compounds and their constituents as the compound’s degree of compositionality. This paper provides a systematic evaluation of vector-space reduction variants across kinds, exploring reductions based on part-of-speech next to and also in combination with Principal Components Analysis using Singular Value and word2vec embeddings. We show that word2vec and nouns only dimensionality reductions are the most successful and stable vector space variants for our task.


Varying Vector Representations and Integrating Meaning Shifts into a PageRank Model for Automatic Term Extraction

Anurag Nigam, Anna Hätty and Sabine Schulte im Walde

We perform a comparative study for automatic term extraction from domain-specific language using a PageRank model with different edge-weighting methods. We vary vector space representations within the PageRank graph algorithm, and we go beyond standard co-occurrence and investigate the influence of measures of association strength and first- vs. second-order co-occurrence. In addition, we incorporate meaning shifts from general to domain-specific language as personalized vectors, in order to distinguish between termhood strengths of ambiguous words across word senses. Our study is performed for two domain-specific English corpora: ACL and do-it-yourself (DIY); and a domain-specific German corpus: cooking. The models are assessed by applying average precision and the roc score as evaluation metrices.


Rigor Mortis: Annotating MWEs with a Gamified Platform

Karën Fort, Bruno Guillaume, Yann-Alan Pilatte, Mathieu Constant and Nicolas Lefèbvre

We present here Rigor Mortis, a gamified crowdsourcing platform designed to evaluate the intuition of the speakers, then train them to annotate multi-word expressions (MWEs) in French corpora. We previously showed that the speakers’ intuition is reasonably good (65% in recall on non-fixed MWE). We detail here the annotation results, after a training phase using some of the tests developed in the PARSEME-FR project.


A Multi-word Expression Dataset for Swedish

Murathan Kurfalı, Robert Östling, Johan Sjons and Mats Wirén

We present a new set of 96 Swedish multi-word expressions annotated with degree of (non-compositionality. In contrast to most previous compositionality datasets we also consider syntactically complex constructions and publish a formal specification of each expression. This allows evaluation of computational models beyond word bigrams, which have so far been the norm. Finally, we use the annotations to evaluate a system for automatic compositionality estimation based on distributional semantics. Our analysis of the disagreements between human annotators and the distributional model reveal interesting questions related to the perception of compositionality, and should be informative to future work in the area.


A Joint Approach to Compound Splitting and Idiomatic Compound Detection

Irina Krotova, Sergey Aksenov and 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%


Dedicated Language Resources for Interdisciplinary Research on Multiword Expressions: Best Thing since Sliced Bread

Ferdy Hubers, Catia Cucchiarini and Helmer Strik

Multiword expressions such as idioms (beat about the bush), collocations (plastic surgery) and lexical bundles (in the middle of) are challenging for disciplines like Natural Language Processing (NLP), psycholinguistics and second language acquisition, , due to their more or less fixed character. Idiomatic expressions are especially problematic, because they convey a figurative meaning that cannot always be inferred from the literal meanings of the component words. Researchers acknowledge that important properties that characterize idioms such as frequency of exposure, familiarity, transparency, and imageability, should be taken into account in research, but these are typically properties that rely on subjective judgments. This is probably one of the reasons why many studies that investigated idiomatic expressions collected limited information about idiom properties for very small numbers of idioms only. In this paper we report on cross-boundary work aimed at developing a set of tools and language resources that are considered crucial for this kind of multifaceted research. We discuss the results of our research and suggest possible avenues for future research


Detecting Multiword Expression Type Helps Lexical Complexity Assessment

Ekaterina Kochmar, Sian Gooding and Matthew Shardlow

Multiword expressions (MWEs) represent lexemes that should be treated as single lexical units due to their idiosyncratic nature. Multiple NLP applications have been shown to benefit from MWE identification, however the research on lexical complexity of MWEs is still an under-explored area. In this work, we re-annotate the Complex Word Identification Shared Task 2018 dataset of Yimam et al. (2017), which provides complexity scores for a range of lexemes, with the types of MWEs. We release the MWE-annotated dataset with this paper, and we believe this dataset represents a valuable resource for the text simplification community. In addition, we investigate which types of expressions are most problematic for native and non-native readers. Finally, we show that a lexical complexity assessment system benefits from the information about MWE types.


Named Entity Recognition

Back to Top


Introducing RONEC – the Romanian Named Entity Corpus

Stefan Daniel Dumitrescu and Andrei-Marius Avram

We present RONEC – the Named Entity Corpus for the Romanian language. The corpus contains over 26000 entities in ~5000 annotated sentences, belonging to 16 distinct classes. The sentences have been extracted from a copy-right free newspaper, covering several styles. This corpus represents the first initiative in the Romanian language space specifically targeted for named entity recognition. It is available in BRAT and CoNLL-U Plus formats, and it is free to use and extend at


A Semi-supervised Approach for De-identification of Swedish Clinical Text

Hanna Berg and Hercules Dalianis

An abundance of electronic health records (EHR) is produced every day within healthcare. The records possess valuable information for research and future improvement of healthcare. Multiple efforts have been done to protect the integrity of patients while making electronic health records usable for research by removing personally identifiable information in patient records. Supervised machine learning approaches for de-identification of EHRs need annotated data for training, annotations that are costly in time and human resources. The annotation costs for clinical text is even more costly as the process must be carried out in a protected environment with a limited number of annotators who must have signed confidentiality agreements. In this paper is therefore, a semi-supervised method proposed, for automatically creating high-quality training data. The study shows that the method can be used to improve recall from 84.75% to 89.20% without sacrificing precision to the same extent, dropping from 95.73% to 94.20%. The model’s recall is arguably more important for de-identification than precision.


A Chinese Corpus for Fine-grained Entity Typing

Chin Lee, Hongliang Dai, Yangqiu Song and Xin Li

Fine-grained entity typing is a challenging task with wide applications. However, most existing datasets for this task are in English. In this paper, we introduce a corpus for Chinese fine-grained entity typing that contains 4,800 mentions manually labeled through crowdsourcing. Each mention is annotated with free-form entity types. To make our dataset useful in more possible scenarios, we also categorize all the fine-grained types into 10 general types. Finally, we conduct experiments with some neural models whose structures are typical in fine-grained entity typing and show how well they perform on our dataset. We also show the possibility of improving Chinese fine-grained entity typing through cross-lingual transfer learning.


Czech Historical Named Entity Corpus v 1.0

Helena Hubková, Pavel Kral and Eva Pettersson

As the number of digitized archival documents increases very rapidly, named entity recognition (NER) in historical documents has become very important for information extraction and data mining. For this task an annotated corpus is needed, which has up to now been missing for Czech. In this paper we present a new annotated data collection for historical NER, composed of Czech historical newspapers. This corpus is freely available for research purposes. For this corpus, we have defined relevant domain-specific named entity types and created an annotation manual for corpus labelling. We further conducted some  experiments on this corpus using recurrent neural networks. We  experimented with randomly initialized embeddings and static and dynamic fastText word embeddings. We achieved 0.73 F1 score with a bidirectional LSTM model using static fastText embeddings.


CodE Alltag 2.0 — A Pseudonymized German-Language Email Corpus

Elisabeth Eder, Ulrike Krieg-Holz and Udo Hahn

The vast amount of social communication distributed over various electronic media channels (tweets, blogs, emails, etc.), so-called user-generated content (UGC), creates entirely new opportunities for today’s NLP research. Yet, data privacy concerns implied by the unauthorized use of these text streams as a data resource are often neglected. In an attempt to reconciliate the diverging needs of unconstrained raw data use and preservation of data privacy in digital communication, we here investigate the automatic recognition of privacy-sensitive stretches of text in UGC and provide an algorithmic solution for the protection of personal data via pseudonymization. Our focus is directed at the de-identification of emails where personally identifying information does not only refer to the sender but also to those people, locations, dates, and other identifiers mentioned in greetings, boilerplates and the content-carrying body of emails. We evaluate several de-identification procedures and systems on two hitherto non-anonymized German-language email corpora (CodE AlltagS+d and CodE AlltagXL), and generate fully pseudonymized versions for both (CodE Alltag 2.0) in which personally identifying information of all social actors addressed in these mails has been camouflaged (to the greatest extent possible).


A Dataset of German Legal Documents for Named Entity Recognition

Elena Leitner, Georg Rehm and 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.


Sensitive Data Detection and Classification in Spanish Clinical Text: Experiments with BERT

Aitor García Pablos, Naiara Perez and Montse Cuadros

Massive digital data processing provides a wide range of opportunities and benefits, but at the cost of endangering personal data privacy. Anonymisation consists in removing or replacing sensitive information from data, enabling its exploitation for different purposes while preserving the privacy of individuals. Over the years, a lot of automatic anonymisation systems have been proposed; however, depending on the type of data, the target language or the availability of training documents, the task remains challenging still. The emergence of novel deep-learning models during the last two years has brought large improvements to the state of the art in the field of Natural Language Processing. These advancements have been most noticeably led by BERT, a model proposed by Google in 2018, and the shared language models pre-trained on millions of documents. In this paper, we use a BERT-based sequence labelling model to conduct a series of anonymisation experiments on several clinical datasets in Spanish. We also compare BERT with other algorithms. The experiments show that a simple BERT-based model with general-domain pre-training obtains highly competitive results without any domain specific feature engineering.


Named Entities in Medical Case Reports: Corpus and Experiments

Sarah Schulz, Jurica Ševa, Samuel Rodriguez, Malte Ostendorff and 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.


Hedwig: A Named Entity Linker

Marcus Klang and Pierre Nugues

Named entity linking is the task of identifying mentions of named things in text, such as “Barack Obama” or “New York”, and linking these mentions to unique identifiers. In this paper, we describe Hedwig, an end-to-end named entity linker, which uses a combination of word and character BILSTM models for mention detection, a Wikidata and Wikipedia-derived knowledge base with global information aggregated over nine language editions, and a PageRank algorithm for entity linking. We evaluated Hedwig on the TAC2017 dataset, consisting of news texts and discussion forums, and we obtained a final score of 59.9% on CEAFmC+, an improvement over our previous generation linker Ugglan, and a trilingual entity link score of 71.9%.


An Experiment in Annotating Animal Species Names from ISTEX Resources

Sabine Barreaux and Dominique Besagni

To exploit scientific publications from global research for TDM purposes, the ISTEX platform enriched its data with value-added information to ease access to its full-text documents. We built an experiment to explore new enrichment possibilities in documents focussing on scientific named entities recognition which could be integrated into ISTEX resources. This led to testing two detection tools for animal species names in a corpus of 100 documents in zoology. This makes it possible to provide the French scientific community with an annotated reference corpus available for use to measure these tools’ performance.


Where are we in Named Entity Recognition from Speech?

Antoine Caubrière, Sophie Rosset, Yannick Estève, Antoine Laurent and Emmanuel Morin

Named entity recognition (NER) from speech is usually made through a pipeline process that consists in (i) processing audio using an automatic speech recognition system (ASR) and (ii) applying a NER to the ASR outputs. The latest data available for named entity extraction from speech in French were produced during the ETAPE evaluation campaign in 2012. Since the publication of ETAPE’s campaign results, major improvements were done on NER and ASR systems, especially with the development of neural approaches for both of these components. In addition, recent studies have shown the capability of End-to-End (E2E) approach for NER / SLU tasks. In this paper, we propose a study of the improvements made in speech recognition and named entity recognition for pipeline approaches. For this type of systems, we propose an original 3-pass approach. We also explore the capability of an E2E system to do structured NER. Finally, we compare the performances of ETAPE’s systems (state-of-the-art systems in 2012) with the performances obtained using current technologies. The results show the interest of the E2E approach, which however remains below an updated pipeline approach.


Tagging Location Phrases in Text

Paul McNamee, James Mayfield, Cash Costello, Caitlyn Bishop and Shelby Anderson

For over thirty years researchers have studied the problem of automatically detecting named entities in written language. Throughout this time the majority of such work has focused on detection and classification of entities into coarse-grained types like: PERSON, ORGANIZATION, and LOCATION. Less attention has been focused on non-named mentions of entities, including non-named location phrases such as “the medical clinic in Telonge” or “2 km below the Dolin Maniche bridge”. In this work we describe the Location Phrase Detection task to identify such spans. Our key accomplishments include: developing a sequential tagging approach; crafting annotation guidelines; building annotated datasets for English and Russian news; and, conducting experiments in automated detection of location phrases with both statistical and neural taggers. This work is motivated by extracting rich location information to support situational awareness during humanitarian crises such as natural disasters.


ScienceExamCER: A High-Density Fine-Grained Science-Domain Corpus for Common Entity Recognition

Hannah Smith, Zeyu Zhang, John Culnan and Peter Jansen

Named entity recognition identifies common classes of entities in text, but these entity labels are generally sparse, limiting utility to downstream tasks.  In this work we present ScienceExamCER, a densely-labeled semantic classification corpus of 133k mentions in the science exam domain where nearly all (96%) of content words have been annotated with one or more fine-grained semantic class labels including taxonomic groups, meronym groups, verb/action groups, properties and values, and synonyms.  Semantic class labels are drawn from a manually-constructed fine-grained typology of 601 classes generated through a data-driven analysis of 4,239 science exam questions.  We show an off-the-shelf BERT-based named entity recognition model modified for multi-label classification achieves an accuracy of 0.85 F1 on this task, suggesting strong utility for downstream tasks in science domain question answering requiring densely-labeled semantic classification.


NorNE: Annotating Named Entities for Norwegian

Fredrik Jørgensen, Tobias Aasmoe, Anne-Stine Ruud Husevåg, Lilja Øvrelid and Erik Velldal

This paper presents NorNE, a manually annotated corpus of named entities which extends the annotation of the existing Norwegian Dependency Treebank. Comprising both of the official standards of written Norwegian (Bokmål and Nynorsk), the corpus contains around 600,000 tokens and annotates a rich set of entity types including persons, organizations, locations, geo-political entities, products, and events, in addition to a class corresponding to nominals derived from names. We here present details on the annotation effort, guidelines, inter-annotator agreement and an experimental analysis of the corpus using a neural sequence labeling architecture.


Tag Me If You Can! Semantic Annotation of Biodiversity Metadata with the QEMP Corpus and the BiodivTagger

Felicitas Löffler, Nora Abdelmageed, Samira Babalou, Pawandeep Kaur and Birgitta König-Ries

Dataset Retrieval is gaining importance due to a large amount of research data and the great demand for reusing scientific data. Dataset Retrieval is mostly based on metadata, structured information about the primary data. Enriching these metadata with semantic annotations based on Linked Open Data (LOD) enables datasets, publications and authors to be connected and expands the search on semantically related terms. In this work, we introduce the BiodivTagger, an ontology-based Information Extraction pipeline, developed for metadata from biodiversity research. The system recognizes biological, physical and chemical processes, environmental terms, data parameters and phenotypes as well as materials and chemical compounds and links them to concepts in dedicated ontologies. To evaluate our pipeline, we created a gold standard of 50 metadata files (QEMP corpus) selected from five different data repositories in biodiversity research. To the best of our knowledge, this is the first annotated metadata corpus for biodiversity research data. The results reveal a mixed picture. While materials and data parameters are properly matched to ontological concepts in most cases, some ontological issues occurred for processes and environmental terms.


Towards a Versatile Medical-Annotation Guideline Feasible Without Heavy Medical Knowledge: Starting From Critical Lung Diseases

Shuntaro Yada, Ayami Joh, Ribeka Tanaka, Fei Cheng, Eiji Aramaki and Sadao Kurohashi

Applying natural language processing (NLP) to medical and clinical texts can bring important social benefits by mining valuable information from unstructured text. A popular application for that purpose is named entity recognition (NER), but the annotation policies of existing clinical corpora have not been standardized across clinical texts of different types. This paper presents an annotation guideline aimed at covering medical documents of various types such as radiography interpretation reports and medical records. Furthermore, the annotation was designed to avoid burdensome requirements related to medical knowledge, thereby enabling corpus development without medical specialists. To achieve these design features, we specifically focus on critical lung diseases to stabilize linguistic patterns in corpora. After annotating around 1100 electronic medical records following the annotation scheme, we demonstrated its feasibility using an NER task. Results suggest that our guideline is applicable to large-scale clinical NLP projects.


Creating a Dataset for Named Entity Recognition in the Archaeology Domain

Alex Brandsen, Suzan Verberne, Milco Wansleeben and Karsten Lambers

In this paper, we present the development of a training dataset for Dutch Named Entity Recognition (NER) in the archaeology domain. This dataset was created as there is a dire need for semantic search within archaeology, in order to allow archaeologists to find structured information in collections of Dutch excavation reports, currently totalling around 60,000 (658 million words) and growing rapidly. To guide this search task, NER is needed. We created rigorous annotation guidelines in an iterative process, then instructed five archaeology students to annotate a number of documents. The resulting dataset contains ~31k annotations between six entity types (artefact, time period, place, context, species & material). The inter-annotator agreement is 0.95, and when we used this data for machine learning, we observed an increase in F1 score from 0.51 to 0.70 in comparison to a machine learning model trained on a dataset created in prior work. This indicates that the data is of high quality, and can confidently be used to train NER classifiers.


Development of a Medical Incident Report Corpus with Intention and Factuality Annotation

Hongkuan Zhang, Ryohei Sasano, Koichi Takeda and Zoie Shui-Yee Wong

Medical incident reports (MIRs) are documents that record what happened in a medical incident. A typical MIR consists of two sections: a structured categorical part and an unstructured text part. Most texts in MIRs describe what medication was intended to be given and what was actually given, because what happened in an incident is largely due to discrepancies between intended and actual medications. Recognizing the intention of clinicians and the factuality of medication is essential to understand the causes of medical incidents and avoid similar incidents in the future. Therefore, we are developing an MIR corpus with annotation of intention and factuality as well as of medication entities and their relations. In this paper, we present our annotation scheme with respect to the definition of medication entities that we take into account, the method to annotate the relations between entities, and the details of the intention and factuality annotation. We then report the annotated corpus consisting of 349 Japanese medical incident reports.


ProGene – A Large-scale, High-Quality Protein-Gene Annotated Benchmark Corpus

Erik Faessler, Luise Modersohn, Christina Lohr and Udo Hahn

Genes and proteins constitute the fundamental entities of molecular genetics. We here introduce ProGene (formerly called  FSU-PRGE), a corpus that reflects our efforts to cope with this important class of named entities within the framework of a long-lasting large-scale annotation campaign at the Jena University Language & Information Engineering (JULIE) Lab. We assembled the entire corpus from 11 subcorpora covering various biological domains to achieve an overall subdomain-independent corpus. It consists of 3,308 MEDLINE abstracts with over 36k sentences and more than 960k tokens annotated with nearly 60k named entity mentions. Two annotators strove for carefully assigning entity mentions to classes of genes/proteins as well as families/groups, complexes, variants and enumerations of those where genes and proteins are represented by a single class. The main purpose of the corpus is to provide a large body of consistent and reliable annotations for supervised training and evaluation of machine learning algorithms in this relevant domain. Furthermore, we provide an evaluation of two state-of-the-art baseline systems — BioBert and flair — on the ProGene corpus. We make the evaluation datasets and the trained models available to encourage comparable evaluations of new methods in the future.


DaNE: A Named Entity Resource for Danish

Rasmus Hvingelby, Amalie Brogaard Pauli, Maria Barrett, Christina Rosted, Lasse Malm Lidegaard and Anders Søgaard

We  present  a  named  entity  annotation  for  the  Danish  Universal  Dependencies treebank using the  CoNLL-2003  annotation  scheme: DaNE. It is the largest publicly available, Danish named entity gold annotation. We evaluate the quality of our annotations intrinsically by double annotating the entire treebank and extrinsically by comparing our annotations to a recently released named entity annotation of the validation and test sections of the Danish Universal Dependencies treebank.  We benchmark the new resource by training and evaluating competitive architectures for supervised named entity recognition (NER), including FLAIR, monolingual (Danish) BERT and multilingual BERT. We explore cross-lingual transfer in multilingual BERT from five related languages in zero-shot and direct transfer setups, and we show that even with our modestly-sized training set, we improve Danish NER over a recent cross-lingual approach, as well as over zero-shot transfer from five related languages.  Using multilingual BERT, we achieve higher performance by fine-tuning on both  DaNE  and  a  larger  Bokmål  (Norwegian)  training set compared to only using  DaNE.  However,  the  highest  performance  isachieved by using a Danish BERT fine-tuned on DaNE. Our dataset enables improvements and applicability for Danish NER beyond cross-lingual methods.  We employ a thorough error analysis of the predictions of the best models for seen and unseen entities, as well as their robustness on un-capitalized text. The annotated dataset and all the trained models are made publicly available.


Fine-grained Named Entity Annotations for German Biographic Interviews

Josef Ruppenhofer, Ines Rehbein and Carolina Flinz

We present a fine-grained NER annotations with 30 labels and apply it to German data. Building on the OntoNotes 5.0 NER inventory, our scheme is adapted for a corpus of transcripts of biographic interviews by adding categories for AGE and LAN(guage) and also features extended numeric and temporal categories. Applying the scheme to the spoken data as well as a collection of teaser tweets from newspaper sites, we can confirm its generality for both domains, also achieving good inter-annotator agreement. We also show empirically how our inventory relates to the well-established 4-category NER inventory by re-annotating a subset of the GermEval 2014 NER coarse-grained dataset with our fine label inventory. Finally, we use a BERT-based system to establish some baseline models for NER tagging on our two new datasets. Global results in in-domain testing are quite high on the two datasets, near what was achieved for the coarse inventory on the CoNLLL2003 data. Cross-domain testing produces much lower results due to the severe domain differences.


A Broad-coverage Corpus for Finnish Named Entity Recognition

Jouni Luoma, Miika Oinonen, Maria Pyykönen, Veronika Laippala and Sampo Pyysalo

We present a new manually annotated corpus for broad-coverage named entity recognition for Finnish. Building on the original Universal Dependencies Finnish corpus of 754 documents (200,000 tokens) representing ten different genres of text,  we introduce annotation marking person, organization, location, product and event names as well as dates. The new annotation identifies in total over 10,000 mentions. An evaluation of inter-annotator agreement indicates that the quality and consistency of annotation are high, at 94.5% F-score for exact match. A comprehensive evaluation using state-of-the-art machine learning methods demonstrates that the new resource maintains compatibility with a previously released single-domain corpus for Finnish NER and makes it possible to recognize named entity mentions in texts drawn from most domains at precision and recall approaching or exceeding 90%. Remaining challenges such as the identification of names in blog posts and transcribed speech are also identified. The newly introduced Turku NER corpus and related resources introduced in this work are released under open licenses via .


Embeddings for Named Entity Recognition in Geoscience Portuguese Literature

Bernardo Consoli, Joaquim Santos, Diogo Gomes, Fabio Cordeiro, Renata Vieira and Viviane Moreira

This work focuses on Portuguese Named Entity Recognition (NER) in the Geology domain. The only domain-specific dataset in the Portuguese language annotated for NER is the GeoCorpus. Our approach relies on BiLSTM-CRF neural networks (a widely used type of network for this area of research) that use vector and tensor embedding representations. Three types of embedding models were used (Word Embeddings, Flair Embeddings, and Stacked Embeddings) under two versions (domain-specific and generalized). The domain specific Flair Embeddings model was originally trained with a generalized context in mind, but was then fine-tuned with domain-specific Oil and Gas corpora, as there simply was not enough domain corpora to properly train such a model. Each of these embeddings was evaluated separately, as well as stacked with another embedding. Finally, we achieved state-of-the-art results for this domain with one of our embeddings, and we performed an error analysis on the language model that achieved the best results. Furthermore, we investigated the effects of domain-specific versus generalized embeddings.


Establishing a New State-of-the-Art for French Named Entity Recognition

Pedro Javier Ortiz Suárez, Yoann Dupont, Benjamin Muller, Laurent Romary and Benoît Sagot

The French TreeBank developed at the University Paris 7 is the main source of morphosyntactic and syntactic annotations for French. However, it does not include explicit information related to named entities, which are among the most useful information for several natural language processing tasks and applications. Moreover, no large-scale French corpus with named entity annotations contain referential information, which complement the type and the span of each mention with an indication of the entity it refers to. We have manually annotated the French TreeBank with such information, after an automatic pre-annotation step. We sketch the underlying annotation guidelines and we provide a few figures about the resulting annotations.


Building OCR/NER Test Collections

Dawn Lawrie, James Mayfield and 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.


Reconstructing NER Corpora: a Case Study on Bulgarian

Iva Marinova, Laska Laskova, Petya Osenova, Kiril Simov and Alexander Popov

The paper reports on the usage of deep learning methods for improving a Named Entity Recognition (NER) training corpus and for predicting and annotating new types in a test corpus. We show how the annotations in a type-based corpus of named entities (NE) were populated as occurrences within it, thus ensuring density of the training information. A deep learning model was adopted for discovering inconsistencies in the initial annotation and for learning new NE types. The evaluation results get improved after data curation, randomization and deduplication.


Natural Language Generation

Back to Top


MucLex: A German Lexicon for Surface Realisation

Kira Klimt, Daniel Braun, Daniela Schneider and Florian Matthes

Language resources for languages other than English are often scarce. Rule-based surface realisers need elaborate lexica in order to be able to generate correct language, especially in languages like German, which include many irregular word forms. In this paper, we present MucLex, a German lexicon for the Natural Language Generation task of surface realisation, based on the crowd-sourced online lexicon Wiktionary. MucLex contains more than 100,000 lemmata and more than 670,000 different word forms in a well-structured XML file and is available under the Creative Commons BY-SA 3.0 license.


Generating Major Types of Chinese Classical Poetry in a Uniformed Framework

Jinyi Hu and Maosong Sun

Poetry generation is an interesting research topic in the field of text generation. As one of the most valuable literary and cultural heritages of China, Chinese classical poetry is very familiar and loved by Chinese people from generation to generation. It has many particular characteristics in its language structure, ranging from form, sound to meaning, thus is regarded as an ideal testing task for text generation. In this paper, we propose a GPT-2 based uniformed framework for generating major types of Chinese classical poems. We define a unified format for formulating all types of training samples by integrating detailed form information, then present a simple form- stressed weighting method in GPT-2 to strengthen the control to the form of the generated poems, with special emphasis on those forms with longer body length. Preliminary experimental results show this enhanced model can generate Chinese classical poems of major types with high quality in both form and content, validating the effectiveness of the proposed strategy. The model has been incorporated into Jiuge, the most influential Chinese classical poetry generation system developed by Tsinghua University.


Video Caption Dataset for Describing Human Actions in Japanese

Yutaro Shigeto, Yuya Yoshikawa, Jiaqing Lin and 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.”


Decode with Template: Content Preserving Sentiment Transfer

Zhiyuan Wen, Jiannong Cao, Ruosong Yang and Senzhang Wang

Sentiment transfer aims to change the underlying sentiment of input sentences. The two major challenges in existing works lie in  (1) effectively disentangling the original sentiment from input sentences; and (2) preserving the semantic content while transferring the sentiment. We find that identifying the sentiment-irrelevant content from input sentences to facilitate generating output sentences could address the above challenges and then propose the Decode with Template model in this paper. We first mask the explicit sentiment words in input sentences and use the rest parts as templates to eliminate the original sentiment. Then, we input the templates and the target sentiments into our bidirectionally guided variational auto-encoder (VAE) model to generate output. In our method, the template preserves most of the semantics in input sentences, and the bidirectionally guided decoding captures both forward and backward contextual information to generate output. Both two parts contribute to better content preservation. We evaluate our method on two review datasets, Amazon and Yelp, with automatic evaluation methods and human rating. The experimental results show that our method significantly outperforms state-of-the-art models, especially in content preservation.


Best Student Forcing: A Simple Training Mechanism in Adversarial Language Generation

Jonathan Sauder, Ting Hu, Xiaoyin Che, Goncalo Mordido, Haojin Yang and 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.


Controllable Sentence Simplification

Louis Martin, Éric de la Clergerie, Benoît Sagot and Antoine Bordes

Text simplification aims at making a text easier to read and understand by simplifying grammar and structure while keeping the underlying information identical. It is often considered an all-purpose generic task where the same simplification is suitable for all; however multiple audiences can benefit from simplified text in different ways. We adapt a discrete parametrization mechanism that provides explicit control on simplification systems based on Sequence-to-Sequence models. As a result, users can condition the simplifications returned by a model on attributes such as length, amount of  paraphrasing, lexical complexity and syntactic complexity. We also show that carefully chosen values of these attributes allow out-of-the-box Sequence-to-Sequence models to outperform their standard counterparts on simplification benchmarks. Our model, which we call ACCESS (as shorthand for AudienCe-CEntric Sentence Simplification), establishes the state of the art at 41.87 SARI  on the WikiLarge test set, a +1.42 improvement over the best previously reported score.


Exploring Transformer Text Generation for Medical Dataset Augmentation

Ali Amin-Nejad, Julia Ive and 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.


Multi-lingual Mathematical Word Problem Generation using Long Short Term Memory Networks with Enhanced Input Features

Vijini Liyanage and Surangika Ranathunga

A Mathematical Word Problem (MWP) differs from a general textual representation due to the fact that it is comprised of numerical quantities and units, in addition to text. Therefore, MWP generation should be carefully handled. When it comes to multi-lingual MWP generation, language specific morphological and syntactic features become additional constraints. Standard template-based MWP generation techniques are incapable of identifying these language specific constraints, particularly in morphologically rich yet low resource languages such as Sinhala and Tamil. This paper presents the use of a Long Short Term Memory (LSTM) network that is capable of generating elementary level MWPs, while satisfying the aforementioned constraints. Our approach feeds a combination of character embeddings, word embeddings, and Part of Speech (POS) tag embeddings to the LSTM, in which attention is provided for numerical values and units. We trained our model for three languages, English, Sinhala and Tamil using separate MWP datasets. Irrespective of the language and the type of the MWP, our model could generate accurate single sentenced and multi sentenced problems. Accuracy reported in terms of average BLEU score for English, Sinhala and Tamil languages were 22.97%, 24.49% and 20.74%, respectively.


Neural Language Representation Models

Back to Top


Time-Aware Word Embeddings for Three Lebanese News Archives

Jad Doughman, Fatima Abu Salem and Shady Elbassuoni

Word embeddings have proven to be an effective method for capturing semantic relations among distinct terms within a large corpus. In this paper, we present a set of word embeddings learnt from three large Lebanese news archives, which collectively consist of 609,386 scanned newspaper images and spanning a total of 151 years, ranging from 1933 till 2011. The diversified ideological nature of the news archives alongside the temporal variability of the embeddings offer a rare glimpse onto the variation of word representation across the left-right political spectrum. To train the word embeddings, Google’s Tesseract 4.0 OCR engine was employed to transcribe the scanned news archives, and various archive-level as well as decade-level word embeddings were learnt. To evaluate the accuracy of the learnt word embeddings, a benchmark of analogy tasks was used. Finally, we demonstrate an interactive system that allows the end user to visualize for a given word of interest, the variation of the top-k closest words in the embedding space as a function of time and across news archives using an animated scatter plot.


GGP: Glossary Guided Post-processing for Word Embedding Learning

Ruosong Yang, Jiannong Cao and Zhiyuan Wen

Word embedding learning is the task to map each word into a low-dimensional and continuous vector based on a large corpus. To enhance corpus based word embedding models, researchers utilize domain knowledge to learn more distinguishable representations via joint optimization and post-processing based models. However, joint optimization based models require much training time. Existing post-processing models mostly consider semantic knowledge while learned embedding models show less functional information. Glossary is a comprehensive linguistic resource. And in previous works, the glossary is usually used to enhance the word representations via joint optimization based methods. In this paper, we post-process pre-trained word embedding models with incorporating the glossary and capture more topical and functional information. We propose GGP (Glossary Guided Post-processing word embedding) model which consists of a global post-processing function to fine-tune each word vector, and an auto-encoding model to learn sense representations, furthermore, constrains each post-processed word representation and the composition of its sense representations to be similar. We evaluate our model by comparing it with two state-of-the-art models on six word topical/functional similarity datasets, and the results show that it outperforms competitors by an average of 4.1% across all datasets. And our model outperforms GloVe by more than 7%.


High Quality ELMo Embeddings for Seven Less-Resourced Languages

Matej Ulčar and Marko Robnik-Šikonja

Recent results show that deep neural networks using contextual embeddings significantly outperform non-contextual embeddings on a majority of text classification task. We offer precomputed embeddings from popular contextual ELMo model for seven languages: Croatian, Estonian, Finnish, Latvian, Lithuanian, Slovenian, and Swedish. We demonstrate that the quality of embeddings strongly depends on the size of training set and show that existing publicly available ELMo embeddings for listed languages shall be improved. We train new ELMo embeddings on much larger training sets and show their advantage over baseline non-contextual FastText embeddings. In evaluation, we use two benchmarks, the analogy task and the NER task.


Is Language Modeling Enough? Evaluating Effective Embedding Combinations

Rudolf Schneider, Tom Oberhauser, Paul Grundmann, Felix Alexander Gers, Alexander Loeser and Steffen Staab

Universal embeddings, such as BERT or ELMo, are useful for a broad set of natural language processing tasks like text classification or sentiment analysis. Moreover, specialized embeddings also exist for tasks like topic modeling or named entity disambiguation. We study if we can complement these universal embeddings with specialized embeddings. We conduct an in-depth evaluation of nine well known natural language understanding tasks with SentEval. Also, we extend SentEval with two additional tasks to the medical domain. We present PubMedSection, a novel topic classification dataset focussed on the biomedical domain. Our comprehensive analysis covers 11 tasks and combinations of six embeddings. We report that combined embeddings outperform state of the art universal embeddings without any embedding fine-tuning. We observe that adding topic model based embeddings helps for most tasks and that differing pre-training tasks encode complementary features. Moreover, we present new state of the art results on the MPQA and SUBJ tasks in SentEval.


Language Modeling with a General Second-Order RNN

Diego Maupomé and Marie-Jean Meurs

Different Recurrent Neural Network (RNN) architectures update their state in different manners as the input sequence is processed. RNNs including a multiplicative interaction between their current state and the current input, second-order ones, show promising performance in language modeling. In this paper, we introduce a second-order RNNs that generalizes existing ones. Evaluating on the Penn Treebank dataset, we analyze how its different components affect its performance in character-lever recurrent language modeling. We perform our experiments controlling the parameter counts of models. We find that removing the first-order terms does not hinder performance. We perform further experiments comparing the effects of the relative size of the state space and the multiplicative interaction space on performance. Our expectation was that a larger states would benefit language models built on longer documents, and larger multiplicative interaction states would benefit ones built on larger input spaces. However, our results suggest that this is not the case and the optimal relative size is the same for both document tokenizations used.


Towards a Gold Standard for Evaluating Danish Word Embeddings

Nina Schneidermann, Rasmus Hvingelby and 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.


Urban Dictionary Embeddings for Slang NLP Applications

Steven Wilson, Walid Magdy, Barbara McGillivray, Kiran Garimella and 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.


Representation Learning for Unseen Words by Bridging Subwords to Semantic Networks

Yeachan Kim, Kang-Min Kim and SangKeun Lee

Pre-trained word embeddings are widely used in various fields. However, the coverage of pre-trained word embeddings only includes words that appeared in corpora where pre-trained embeddings are learned. It means that the words which do not appear in training corpus are ignored in tasks, and it could lead to the limited performance of neural models. In this paper, we propose a simple yet effective method to represent out-of-vocabulary (OOV) words. Unlike prior works that solely utilize subword information or knowledge, our method makes use of both information to represent OOV words. To this end, we propose two stages of representation learning. In the first stage, we learn subword embeddings from the pre-trained word embeddings by using an additive composition function of subwords. In the second stage, we map the learned subwords into semantic networks (e.g., WordNet). We then re-train the subword embeddings by using lexical entries on semantic lexicons that could include newly observed subwords. This two-stage learning makes the coverage of words broaden to a great extent. The experimental results clearly show that our method provides consistent performance improvements over strong baselines that use subwords or lexical resources separately.


Give your Text Representation Models some Love: the Case for Basque

Rodrigo Agerri, Iñaki San Vicente, Jon Ander Campos, Ander Barrena, Xabier Saralegi, Aitor Soroa and Eneko Agirre

Word embeddings and pre-trained language models allow to build rich representations of text and have enabled improvements across most NLP tasks. Unfortunately they are very expensive to train, and many small companies and research groups tend to use models that have been pre-trained and made available by third parties, rather than building their own. This is suboptimal as, for many languages, the models have been trained on smaller (or lower quality) corpora. In addition, monolingual pre-trained models for non-English languages are not always available. At best, models for those languages are included in multilingual versions, where each language shares the quota of substrings and parameters with the rest of the languages. This is particularly true for smaller languages such as Basque. In this paper we show that a number of monolingual models (FastText word embeddings, FLAIR and BERT language models) trained with larger Basque corpora produce much better results than publicly available versions in downstream NLP tasks, including topic classification, sentiment classification, PoS tagging and NER. This work sets a new state-of-the-art in those tasks for Basque. All benchmarks and models used in this work are publicly available.


On the Correlation of Word Embedding Evaluation Metrics

François Torregrossa, Vincent Claveau, Nihel Kooli, Guillaume Gravier and Robin Allesiardo

Word embeddings intervene in a wide range of natural language processing tasks. These geometrical representations are easy to manipulate for automatic systems. Therefore, they quickly invaded all areas of language processing. While they surpass all predecessors, it is still not straightforward why and how they do so. In this article, we propose to investigate all kind of evaluation metrics on various datasets in order to discover how they correlate with each other. Those correlations lead to 1) a fast solution to select the best word embeddings among many others, 2) a new criterion that may improve the current state of static Euclidean word embeddings, and 3) a way to create a set of complementary datasets, i.e. each dataset quantifies a different aspect of word embeddings.


CBOW-tag: a Modified CBOW Algorithm for Generating Embedding Models from Annotated Corpora

Attila Novák, László Laki and Borbála Novák

In this paper, we present a modified version of the CBOW algorithm implemented in the fastText framework. Our modified algorithm, CBOW-tag builds a vector space model that includes the representation of the original word forms and their annotation at the same time. We illustrate the results by presenting a model built from a corpus that includes morphological and syntactic annotations. The simultaneous presence of unannotated elements and different annotations at the same time in the model makes it possible to constrain nearest neighbour queries to specific types of elements. The model can thus efficiently answer questions such as What do we eat?, What can we do with a skeleton? What else do we do with what we eat?, etc. Error analysis reveals that the model can highlight errors introduced into the annotation by the tagger and parser we used to generate the annotations as well as lexical peculiarities in the corpus itself, especially if we do not limit the vocabulary of the model to frequent items.


Much Ado About Nothing – Identification of Zero Copulas in Hungarian Using an NMT Model

Andrea Dömötör, Zijian Győző Yang and Attila Novák

The research presented in this paper concerns zero copulas in Hungarian, i.e. the phenomenon that nominal predicates lack an explicit verbal copula in the default present tense 3rd person indicative case. We created a tool based on the state-of-the-art transformer architecture implemented in Marian NMT framework that can identify and mark the location of zero copulas, i.e. the position where an overt copula would appear in the non-default cases. Our primary aim was to support quantitative corpus-based linguistic research by creating a tool that can be used to compile a corpus of significant size containing examples of nominal predicates including the location of the zero copulas. We created the training corpus for our system transforming sentences containing overt copulas into ones containing zero copula labels. However, we first needed to disambiguate occurrences of the massively ambiguous verb van `exist/be/have’. We performed this using a rule-base classifier relying on English translations in the English-Hungarian parallel subcorpus of the OpenSubtitles corpus. We created several NMT-based models using different sampling methods and optionally using our baseline model to synthesize additional training data. Our best model obtains almost 90% precision and 80% recall on an in-domain test set.


Leveraging Contextual Embeddings for Detecting Diachronic Semantic Shift

Matej Martinc, Petra Kralj Novak and Senja Pollak

We propose a new method that leverages contextual embeddings for the task of diachronic semantic shift detection by generating time specific word representations from BERT embeddings. The results of our experiments in the domain specific LiverpoolFC corpus suggest that the proposed method has performance comparable to the current state-of-the-art without requiring any time consuming domain adaptation on large corpora. The results on the newly created Brexit news corpus suggest that the method can be successfully used for the detection of a short-term yearly semantic shift. And lastly, the model also shows promising results in a multilingual settings, where the task was to detect differences and similarities between diachronic semantic shifts in different languages.


Improving NMT Quality Using Terminology Injection

Duane K. Dougal and Deryle Lonsdale

Many organizations use domain- or organization-specific words and phrases.  This paper explores the use of vetted terminology as an input to neural machine translation (NMT) for improved results: ensuring that the translation of individual terms is consistent with an approved multilingual terminology collection. We discuss, implement, and evaluate a method for injecting terminology and for evaluating terminology injection. Our use of the long short-term memory (LSTM) attention mechanism prevalent in state-of-the-art NMT systems involves attention vectors for correctly identifying semantic entities and aligning the tokens that represent them, both in the source and the target languages. Appropriate terminology is then injected into matching alignments during decoding. We also introduce a new translation metric more sensitive to approved terminological content in MT output.


Word Embedding Evaluation in Downstream Tasks and Semantic Analogies

Joaquim Santos, Bernardo Consoli and Renata Vieira

Language Models have long been a prolific area of study in the field of Natural Language Processing (NLP). One of the newer kinds of language models, and some of the most used, are Word Embeddings (WE). WE are vector space representations of a vocabulary learned by a non-supervised neural network based on the context in which words appear. WE have been widely used in downstream tasks in many areas of study in NLP. These areas usually use these vector models as a feature in the processing of textual data. This paper presents the evaluation of newly released WE models for the Portuguese langauage, trained with a corpus composed of 4.9 billion tokens. The first evaluation presented an intrinsic task in which WEs had to correctly build semantic and syntactic relations. The second evaluation presented an extrinsic task in which the WE models were used in two downstream tasks: Named Entity Recognition and Semantic Similarity between Sentences. Our results show that a diverse and comprehensive corpus can often outperform a larger, less textually diverse corpus, and that batch training may cause quality loss in WE models.


Detection of Reading Absorption in User-Generated Book Reviews: Resources Creation and Evaluation

Piroska Lendvai, Sándor Darányi, Christian Geng, Moniek Kuijpers, Oier Lopez de Lacalle, Jean-Christophe Mensonides, Simone Rebora and Uwe Reichel

To detect how and when readers are experiencing engagement with a literary work, we bring together empirical literary studies and language technology via focusing on the affective state of absorption. The goal of our resource development is to enable the detection of different levels of reading absorption in millions of user-generated reviews hosted on social reading platforms. We present a corpus of social book reviews in English that we annotated with reading absorption categories. Based on these data, we performed supervised, sentence level, binary classification of the explicit presence vs. absence of the mental state of absorption. We compared the performances of classical machine learners where features comprised sentence representations obtained from a pretrained embedding model (Universal Sentence Encoder) vs. neural classifiers in which sentence embedding vector representations are adapted or fine-tuned while training for the absorption recognition task. We discuss the challenges in creating the labeled data as well as the possibilities for releasing a benchmark corpus.


Ontologies and Wordnet

Back to Top


Developing an Arabic Infectious Disease Ontology to Include Non-Standard Terminology

Lama Alsudias and Paul Rayson

Building ontologies is a crucial part of the semantic web endeavour. In recent years, research interest has grown rapidly in supporting languages such as Arabic in NLP in general but there has been very little research on medical ontologies for Arabic. We present a new Arabic ontology in the infectious disease domain to support various important applications including the monitoring of infectious disease spread via social media. This ontology meaningfully integrates the scientific vocabularies of infectious diseases with their informal equivalents. We use ontology learning strategies with manual checking to build the ontology. We applied three statistical methods for term extraction from selected Arabic infectious diseases articles: TF-IDF, C-value, and YAKE. We also conducted a study, by consulting around 100 individuals, to discover the informal terms related to infectious diseases in Arabic. In future work, we will automatically extract the relations for infectious disease concepts but for now these are manually created. We report two complementary experiments to evaluate the ontology. First, a quantitative evaluation of the term extraction results and an additional qualitative evaluation by a domain expert.


Aligning Wikipedia 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.


The MWN.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 and 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.

Ontology-Style Relation Annotation: A Case Study

Savong Bou, Naoki Suzuki, Makoto Miwa and Yutaka Sasaki

This paper proposes an Ontology-Style Relation (OSR) annotation approach. In conventional Relation Extraction (RE) datasets, relations are annotated as links between entity mentions. In contrast, in our OSR annotation, a relation is annotated as a relation mention (i.e., not a link but a node) and domain and range links are annotated from the relation mention to its argument entity mentions. We expect the following benefits: (1) the relation annotations can be easily converted to Resource Description Framework (RDF) triples to populate an Ontology, (2) some part of conventional RE tasks can be tackled as Named Entity Recognition (NER) tasks. The relation classes are limited to several RDF properties such as domain, range, and subClassOf, and (3) OSR annotations can be clear documentations of Ontology contents. As a case study, we converted an in-house corpus of Japanese traffic rules in conventional annotations into the OSR annotations and built a novel OSR-RoR (Rules of the Road) corpus. The inter-annotator agreements of the conversion were 85-87%. We evaluated the performance of neural NER and RE tools on the conventional and OSR annotations. The experimental results showed that the OSR annotations make the RE task easier while introducing slight complexity into the NER task.


The Ontology of Bulgarian 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 /ѫ/ under stress, the reflex of the back er vowel /ъ/ under stress, and the reflex of the yat vowel /ѣ/ 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.


Spatial AMR: Expanded Spatial Annotation in the Context of a Grounded Minecraft Corpus

Julia Bonn, Martha Palmer, Zheng Cai and Kristin Wright-Bettner

This paper presents an expansion to the Abstract Meaning Representation (AMR) annotation schema that captures fine-grained semantically and pragmatically derived spatial information in grounded corpora. We describe a new lexical category conceptualization and set of spatial annotation tools built in the context of a multimodal corpus consisting of 170 3D structure-building dialogues between a human architect and human builder in Minecraft. Minecraft provides a particularly beneficial spatial relation-elicitation environment because it automatically tracks locations and orientations of objects and avatars in the space according to an absolute Cartesian coordinate system. Through a two-step process of sentence-level and document-level annotation designed to capture implicit information, we leverage these coordinates and bearings in the AMRs in combination with spatial framework annotation to ground the spatial language in the dialogues to absolute space.


English WordNet Random Walk Pseudo-Corpora

Filip Klubička, Alfredo Maldonado, Abhijit Mahalunkar and John Kelleher

This is a resource description paper that describes the creation and properties of a set of pseudo-corpora generated artificially from a random walk over the English WordNet taxonomy. Our WordNet taxonomic random walk implementation allows the exploration of different random walk hyperparameters and the generation of a variety of different pseudo-corpora. We find that different combinations of parameters result in varying statistical properties of the generated pseudo-corpora. We have published a total of 81 pseudo-corpora that we have used in our previous research, but have not exhausted all possible combinations of hyperparameters, which is why we have also published a codebase that allows the generation of additional WordNet taxonomic pseudo-corpora as needed. Ultimately, such pseudo-corpora can be used to train taxonomic word embeddings, as a way of transferring taxonomic knowledge into a word embedding space.


On the Formal Standardization of Terminology Resources: The Case Study of TriMED

Federica Vezzani and Giorgio Maria Di Nunzio

The process of standardization plays an important role in the management of terminological resources. In this context, we present the work of re-modeling an existing multilingual terminological database for the medical domain, named TriMED. This resource was conceived in order to tackle some problems related to the complexity of medical terminology and to respond to different users’ needs. We provide a methodology that should be followed in order to make a termbase compliant to the three most recent ISO/TC 37 standards. In particular, we focus on the definition of i) the structural meta-model of the resource, ii) the data categories provided, and iii) the TBX format for its implementation. In addition to the formal standardization of the resource, we describe the realization of a new data category repository for the management of the TriMED terminological data and a Web application that can be used to access the multilingual terminological records.


Opinion Mining, Sentiment Analysis

Back to Top


Metaphorical Expressions in Automatic Arabic Sentiment Analysis

Israa Alsiyat and 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.


HotelRec: a Novel Very Large-Scale Hotel Recommendation Dataset

Diego Antognini and Boi Faltings

Today, recommender systems are an inevitable part of everyone’s daily digital routine and are present on most internet platforms. State-of-the-art deep learning-based models require a large number of data to achieve their best performance. Many datasets fulfilling this criterion have been proposed for multiple domains, such as Amazon products, restaurants, or beers. However, works and datasets in the hotel domain are limited: the largest hotel review dataset is below the million samples. Additionally, the hotel domain suffers from a higher data sparsity than traditional recommendation datasets and therefore, traditional collaborative-filtering approaches cannot be applied to such data. In this paper, we propose HotelRec, a very large-scale hotel recommendation dataset, based on TripAdvisor, containing 50 million reviews. To the best of our knowledge, HotelRec is the largest publicly available dataset in the hotel domain (50M versus 0.9M) and additionally, the largest recommendation dataset in a single domain and with textual reviews (50M versus 22M). We release HotelRec for further research:


Doctor Who? Framing Through Names and Titles in German

Esther van den Berg, Katharina Korfhage, Josef Ruppenhofer, Michael Wiegand and 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.


Adapt or Get Left Behind: Domain Adaptation through BERT Language Model Finetuning for Aspect-Target Sentiment Classification

Alexander Rietzler, Sebastian Stabinger, Paul Opitz and Stefan Engl

Aspect-Target Sentiment Classification (ATSC) is a subtask of Aspect-Based Sentiment Analysis (ABSA), which has many applications e.g. in e-commerce, where data and insights from reviews can be leveraged to create value for businesses and customers. Recently, deep transfer-learning methods have been applied successfully to a myriad of Natural Language Processing (NLP) tasks, including ATSC. Building on top of the prominent BERT language model, we approach ATSC using a two-step procedure: self-supervised domain-specific BERT language model finetuning, followed by supervised task-specific finetuning. Our findings on how to best exploit domain-specific language model finetuning enable us to produce new state-of-the-art performance on the SemEval 2014 Task 4 restaurants dataset. In addition, to explore the real-world robustness of our models, we perform cross-domain evaluation. We show that a cross-domain adapted BERT language model performs significantly better than strong baseline models like vanilla BERT-base and XLNet-base. Finally, we conduct a case study to interpret model prediction errors.


An Empirical Examination of Online Restaurant Reviews

Hyun Jung Kang and 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.


Manovaad: A Novel Approach to Event Oriented Corpus Creation Capturing Subjectivity and Focus

Lalitha Kameswari and Radhika Mamidi

In today’s era of globalisation, the increased outreach for every event across the world has been leading to conflicting opinions, arguments and disagreements, often reflected in print media and online social platforms. It is necessary to distinguish factual observations from personal judgements in news, as subjectivity in reporting can influence the audience’s perception of reality. Several studies conducted on the different styles of reporting in journalism are essential in understanding phenomena such as media bias and multiple interpretations of the same event. This domain finds applications in fields such as Media Studies, Discourse Analysis, Information Extraction, Sentiment Analysis, and Opinion Mining. We present an event corpus Manovaad-v1.0 consisting of 1035 news articles corresponding to 65 events from 3 levels of newspapers viz., Local, National, and International levels. Using this novel format, we correlate the trends in the degree of subjectivity with the geographical closeness of reporting using a Bi-RNN model. We also analyse the role of background and focus in event reporting and capture the focus shift patterns within a global discourse structure for an event. We do this across different levels of reporting and compare the results with the existing work on discourse processing.


Toward Qualitative Evaluation of Embeddings for Arabic Sentiment Analysis

Amira Barhoumi, Nathalie Camelin, Chafik Aloulou, Yannick Estève and Lamia Hadrich Belguith

In this paper, we propose several protocols to evaluate specific embeddings for Arabic sentiment analysis (SA) task. In fact, Arabic language is characterized by its agglutination and morphological richness contributing to great sparsity that could affect embedding quality. This work presents a study that compares embeddings based on words and lemmas in SA frame. We propose  first to study the evolution of embedding models trained with different types of corpora (polar and non polar) and explore the variation between embeddings by observing the sentiment stability of neighbors in embedding spaces. Then, we evaluate embeddings with a neural architecture based on convolutional neural network (CNN). We make available our pre-trained embeddings to Arabic NLP research community with free to use. We provide also for free resources used to evaluate our embeddings. Experiments are done on the Large Arabic-Book Reviews (LABR) corpus in binary (positive/negative) classification frame. Our best result reaches 91.9%, that is higher than the best previous published one (91.5%).


Annotating Perspectives on Vaccination

Roser Morante, Chantal van Son, Isa Maks and 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.


Aspect On: an Interactive Solution for Post-Editing the Aspect Extraction based on Online Learning

Mara Chinea-Rios, Marc Franco-Salvador and Yassine Benajiba

The task of aspect extraction is an important component of aspect-based sentiment analysis. However, it usually requires an expensive human post-processing to ensure quality. In this work we introduce Aspect On, an interactive solution based on online learning that allows users to post-edit the aspect extraction with little effort. The Aspect On interface shows the aspects extracted by a neural model and, given a dataset, annotates its words with the corresponding aspects. Thanks to the online learning, Aspect On updates the model automatically and continuously improves the quality of the aspects displayed to the user. Experimental results show that Aspect On dramatically reduces the number of user clicks and effort required to post-edit the aspects extracted by the model.


Recommendation Chart of Domains for Cross-Domain Sentiment Analysis: Findings of A 20 Domain Study

Akash Sheoran, Diptesh Kanojia, Aditya Joshi and Pushpak Bhattacharyya

Cross-domain sentiment analysis (CDSA) helps to address the problem of data scarcity in scenarios where labelled data for a domain (known as the target domain) is unavailable or insufficient. However, the decision to choose a domain (known as the source domain) to leverage from is, at best, intuitive. In this paper, we investigate text similarity metrics to facilitate source domain selection for CDSA. We report results on 20 domains (all possible pairs) using 11 similarity metrics. Specifically, we compare CDSA performance with these metrics for different domain-pairs to enable the selection of a suitable source domain, given a target domain. These metrics include two novel metrics for evaluating domain adaptability to help source domain selection of labelled data and utilize word and sentence-based embeddings as metrics for unlabelled data. The goal of our experiments is a recommendation chart that gives the K best source domains for CDSA for a given target domain. We show that the best K source domains returned by our similarity metrics have a precision of over 50%, for varying values of K.


Inference Annotation of a Chinese Corpus for Opinion Mining

Liyun Yan, Danni E, Mei Gan, Cyril Grouin and Mathieu Valette

Polarity classification (positive, negative or neutral opinion detection) is well developed in the field of opinion mining. However, existing tools, which perform with high accuracy on short sentences and explicit expressions, have limited success interpreting narrative phrases and inference contexts. In this article, we will discuss an important aspect of opinion mining: inference. We will give our definition of inference, classify different types, provide an annotation framework and analyze the annotation results. While inferences are often studied in the field of Natural-language understanding (NLU), we propose to examine inference as it relates to opinion mining. Firstly, based on linguistic analysis, we clarify what kind of sentence contains an inference. We define five types of inference: logical inference, pragmatic inference, lexical inference, enunciative inference and discursive inference. Second, we explain our annotation framework which includes both inference detection and opinion mining. In short, this manual annotation determines whether or not a target contains an inference. If so, we then define inference type, polarity and topic. Using the results of this annotation, we observed several correlation relations which will be used to determine distinctive features for automatic inference classification in further research. We also demonstrate the results of three preliminary classification experiments.


Cooking Up a Neural-based Model for Recipe Classification

Elham Mohammadi, Nada Naji, Louis Marceau, Marc Queudot, Eric Charton, Leila Kosseim and Marie-Jean Meurs

In this paper, we propose a neural-based model to address the first task of the DEFT 2013 shared task, with the main challenge of a highly imbalanced dataset, using state-of-the-art embedding approaches and deep architectures. We report on our experiments on the use of linguistic features, extracted by Charton et. al. (2014), in different neural models utilizing pretrained embeddings. Our results show that all of the models that use linguistic features outperform their counterpart models that only use pretrained embeddings. The best performing model uses pretrained CamemBERT embeddings as input and CNN as the hidden layer, and uses additional linguistic features. Adding the linguistic features to this model improves its performance by 4.5% and 11.4% in terms of micro and macro F1 scores, respectively, leading to state-of-the-art results and an improved classification of the rare classes.


Enhancing a Lexicon of Polarity Shifters through the Supervised Classification of Shifting Directions

Marc Schulder, Michael Wiegand and Josef Ruppenhofer

The sentiment polarity of an expression (whether it is perceived as positive, negative or neutral) can be influenced by a number of phenomena, foremost among them negation. Apart from closed-class negation words like “no”, “not” or “without”, negation can also be caused by so-called polarity shifters. These are content words, such as verbs, nouns or adjectives, that shift polarities in their opposite direction, e.g. “abandoned” in “abandoned hope” or “alleviate” in “alleviate pain”. Many polarity shifters can affect both positive and negative polar expressions, shifting them towards the opposing polarity. However, other shifters are restricted to a single shifting direction. “Recoup” shifts negative to positive in “recoup your losses”, but does not affect the positive polarity of “fortune” in “recoup a fortune”. Existing polarity shifter lexica only specify whether a word can, in general, cause shifting, but they do not specify when this is limited to one shifting direction. To address this issue we introduce a supervised classifier that determines the shifting direction of shifters. This classifier uses both resource-driven features, such as WordNet relations, and data-driven features like in-context polarity conflicts. Using this classifier we enhance the largest available polarity shifter lexicon.


Dataset Creation and Evaluation of Aspect Based Sentiment Analysis in Telugu, a Low Resource Language

Yashwanth Reddy Regatte, Rama Rohit Reddy Gangula and 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.


A Fine-grained Sentiment Dataset for Norwegian

Lilja Øvrelid, Petter Mæhlum, Jeremy Barnes and 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.


The Design and Construction of a Chinese Sarcasm Dataset

Xiaochang Gong, Qin Zhao, Jun Zhang, Ruibin Mao and 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.


Target-based Sentiment Annotation in Chinese Financial News

Chaofa Yuan, Yuhan Liu, Rongdi Yin, Jun Zhang, Qinling Zhu, Ruibin Mao and 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.


Multi-domain Tweet Corpora for Sentiment Analysis: Resource Creation and Evaluation

Mamta ., Asif Ekbal, Pushpak Bhattacharyya, Shikha Srivastava, Alka Kumar and Tista Saha

Due to the phenomenal growth of online content in recent time, sentiment analysis has attracted attention of the researchers and developers. A number of benchmark annotated corpora are available for domains like movie reviews, product reviews, hotel reviews, etc.The pervasiveness of social media has also lead to a huge amount of content posted by users who are misusing the power of social media to spread false beliefs and to negatively influence others. This type of content is coming from the domains like terrorism, cybersecurity, technology, social issues, etc. Mining of opinions from these domains is important to create a socially intelligent system to provide security to the public and to maintain the law and order situations.  To the best of our knowledge, there is no publicly available tweet corpora for such pervasive domains.  Hence, we firstly create a multi-domain tweet sentiment corpora and then establish a deep neural network based baseline framework to address the above mentioned issues. Annotated corpus has Cohen’s Kappa measurement for annotation quality of 0.770, which shows that the data is of acceptable quality. We are able to achieve 84.65% accuracy for sentiment analysis by using an ensemble of Convolutional Neural Network (CNN), Long Short Term Memory (LSTM), and Gated Recurrent Unit(GRU).


Reproduction and Revival of the Argument Reasoning Comprehension Task

João António Rodrigues, Ruben Branco, João Silva and 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.


Design and Evaluation of SentiEcon: a fine-grained Economic/Financial Sentiment Lexicon from a Corpus of Business News

Antonio Moreno-Ortiz, Javier Fernandez-Cruz and Chantal Pérez Chantal Hernández

In this paper we present, describe, and evaluate SentiEcon, a large, comprehensive, domain-specific computational lexicon designed for sentiment analysis applications, for which we compiled our own corpus of online business news. SentiEcon was created as a plug-in lexicon for the sentiment analysis tool Lingmotif, and thus it follows its data structure requirements and presupposes the availability of a general-language core sentiment lexicon that covers non-specific sentiment-carrying terms and phrases. It contains 6,470 entries, both single and multi-word expressions, each with tags denoting their semantic orientation and intensity. We evaluate SentiEcon’s performance by comparing results in a sentence classification task using exclusively sentiment words as features. This sentence dataset was extracted from business news texts, and included certain key words known to recurrently convey strong semantic orientation, such as “debt”, “inflation” or “markets”. The results show that performance is significantly improved when adding SentiEcon to the general-language sentiment lexicon.


ParlVote: A Corpus for Sentiment Analysis of Political Debates

Gavin Abercrombie and 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.


Offensive Language Detection Using Brown Clustering

Zuoyu Tian and Sandra Kübler

In this study, we investigate the use of  Brown clustering for offensive language detection. Brown clustering has been shown to be of little use when the task involves distinguishing word polarity in sentiment analysis tasks. In contrast to previous work, we train Brown clusters separately on positive and negative sentiment data, but then combine the information into a single complex feature per word. This way of representing words results in stable improvements in offensive language detection, when used as the only features or in combination with words or character n-grams.  Brown clusters  add important information, even when combined with words or character n-grams or with standard word embeddings in a convolutional neural network. However, we also found different trends between the two offensive language data sets we used.


Annotating for Hate Speech: The MaNeCo Corpus and Some Input from Critical Discourse Analysis

Stavros Assimakopoulos, Rebecca Vella Muskat, Lonneke van der Plas and Albert Gatt

This paper presents a novel scheme for the annotation of hate speech in corpora of Web 2.0 commentary. The proposed scheme is motivated by the critical analysis of posts made in reaction to news reports on the Mediterranean migration crisis and LGBTIQ+ matters in Malta, which was conducted under the auspices of the EU-funded C.O.N.T.A.C.T. project. Based on the realisation that hate speech is not a clear-cut category to begin with, appears to belong to a continuum of discriminatory discourse and is often realised through the use of indirect linguistic means, it is argued that annotation schemes for its detection should refrain from directly including the label ‘hate speech,’ as different annotators might have different thresholds as to what constitutes hate speech and what not. In view of this, we propose a multi-layer annotation scheme, which is pilot-tested against a binary ±hate speech classification and appears to yield higher inter-annotator agreement. Motivating the postulation of our scheme, we then present the MaNeCo corpus on which it will eventually be used; a substantial corpus of on-line newspaper comments spanning 10 years.


Marking Irony Activators in a Universal Dependencies Treebank: The Case of an Italian Twitter Corpus

Alessandra Teresa Cignarella, Manuela Sanguinetti, Cristina Bosco and Paolo Rosso

The recognition of irony is a challenging task in the domain of Sentiment Analysis, and the availability of annotated corpora may be crucial for its automatic processing. In this paper we describe a fine-grained annotation scheme centered on irony, in which we highlight the tokens that are responsible for its activation, (irony activators) and their morpho-syntactic features. As our case study we therefore introduce a recently released Universal Dependencies treebank for Italian which includes ironic tweets: TWITTIRÒ-UD. For the purposes of this study, we enriched the existing annotation in the treebank, with a further level that includes irony activators. A description and discussion of the annotation scheme is provided with a definition of irony activators and the guidelines for their annotation. This qualitative study on the different layers of annotation applied on the same dataset can shed some light on the process of human annotation, and irony annotation in particular, and on the usefulness of this representation for developing computational models of irony to be used for training purposes.


HAHA 2019 Dataset: A Corpus for Humor Analysis in Spanish

Luis Chiruzzo, Santiago Castro and Aiala Rosá

This paper presents the development of a corpus of 30,000 Spanish tweets that were crowd-annotated with humor value and funniness score. The corpus contains approximately 38.6% of humorous tweets with an average score of 2.04 in a scale from 1 to 5 for the humorous tweets. The corpus has been used in an automatic humor recognition and analysis competition, obtaining encouraging results from the participants.


Offensive Language Identification in Greek

Zesis Pitenis, Marcos Zampieri and 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.

  • LREC 2020 Paper Dissemination (1/10)
  • LREC 2020 Paper Dissemination (2/10)
  • LREC 2020 Paper Dissemination (3/10)
  • LREC 2020 Paper Dissemination (4/10)
  • LREC 2020 Paper Dissemination (5/10)
  • LREC 2020 Paper Dissemination (6/10)
  • LREC 2020 Paper Dissemination (7/10)
  • LREC 2020 Paper Dissemination (8/10)
  • LREC 2020 Paper Dissemination (9/10)
  • LREC 2020 Paper Dissemination (10/10)