2021 |
Shardul Suryawanshi; Bharathi Raja Chakravarthi Findings of the Shared Task on Troll Meme Classification in Tamil Inproceedings Proceedings of the First Workshop on Speech and Language Technologies for Dravidian Languages, pp. 126–132, Association for Computational Linguistics, Kyiv, 2021. @inproceedings{suryawanshi-chakravarthi-2021-findings, title = {Findings of the Shared Task on Troll Meme Classification in Tamil}, author = {Shardul Suryawanshi and Bharathi Raja Chakravarthi}, url = {https://www.aclweb.org/anthology/2021.dravidianlangtech-1.16}, year = {2021}, date = {2021-04-01}, booktitle = {Proceedings of the First Workshop on Speech and Language Technologies for Dravidian Languages}, pages = {126--132}, publisher = {Association for Computational Linguistics}, address = {Kyiv}, abstract = {The internet has facilitated its user-base with a platform to communicate and express their views without any censorship. On the other hand, this freedom of expression or free speech can be abused by its user or a troll to demean an individual or a group. Demeaning people based on their gender, sexual orientation, religious believes or any other characteristics --trolling-- could cause great distress in the online community. Hence, the content posted by a troll needs to be identified and dealt with before causing any more damage. Amongst all the forms of troll content, memes are most prevalent due to their popularity and ability to propagate across cultures. A troll uses a meme to demean, attack or offend its targetted audience. In this shared task, we provide a resource (TamilMemes) that could be used to train a system capable of identifying a troll meme in the Tamil language. In our TamilMemes dataset, each meme has been categorized into either a ``troll'' or a ``not_troll'' class. Along with the meme images, we also provided the Latin transcripted text from memes. We received 10 system submissions from the participants which were evaluated using the weighted average F1-score. The system with the weighted average F1-score of 0.55 secured the first rank.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } The internet has facilitated its user-base with a platform to communicate and express their views without any censorship. On the other hand, this freedom of expression or free speech can be abused by its user or a troll to demean an individual or a group. Demeaning people based on their gender, sexual orientation, religious believes or any other characteristics --trolling-- could cause great distress in the online community. Hence, the content posted by a troll needs to be identified and dealt with before causing any more damage. Amongst all the forms of troll content, memes are most prevalent due to their popularity and ability to propagate across cultures. A troll uses a meme to demean, attack or offend its targetted audience. In this shared task, we provide a resource (TamilMemes) that could be used to train a system capable of identifying a troll meme in the Tamil language. In our TamilMemes dataset, each meme has been categorized into either a ``troll'' or a ``not_troll'' class. Along with the meme images, we also provided the Latin transcripted text from memes. We received 10 system submissions from the participants which were evaluated using the weighted average F1-score. The system with the weighted average F1-score of 0.55 secured the first rank. |
Bharathi Raja Chakravarthi; Ruba Priyadharshini; Shubhanker Banerjee; Richard Saldanha; John P McCrae; Anand Kumar M; Parameswari Krishnamurthy; Melvin Johnson Findings of the Shared Task on Machine Translation in Dravidian languages Inproceedings Proceedings of the First Workshop on Speech and Language Technologies for Dravidian Languages, pp. 119–125, Association for Computational Linguistics, Kyiv, 2021. @inproceedings{chakravarthi-etal-2021-findings, title = {Findings of the Shared Task on Machine Translation in Dravidian languages}, author = {Bharathi Raja Chakravarthi and Ruba Priyadharshini and Shubhanker Banerjee and Richard Saldanha and John P McCrae and Anand Kumar M and Parameswari Krishnamurthy and Melvin Johnson}, url = {https://www.aclweb.org/anthology/2021.dravidianlangtech-1.15}, year = {2021}, date = {2021-04-01}, booktitle = {Proceedings of the First Workshop on Speech and Language Technologies for Dravidian Languages}, pages = {119--125}, publisher = {Association for Computational Linguistics}, address = {Kyiv}, abstract = {This paper presents an overview of the shared task on machine translation of Dravidian languages. We presented the shared task results at the EACL 2021 workshop on Speech and Language Technologies for Dravidian Languages. This paper describes the datasets used, the methodology used for the evaluation of participants, and the experiments' overall results. As a part of this shared task, we organized four sub-tasks corresponding to machine translation of the following language pairs: English to Tamil, English to Malayalam, English to Telugu and Tamil to Telugu which are available at https://competitions.codalab.org/competitions/27650. We provided the participants with training and development datasets to perform experiments, and the results were evaluated on unseen test data. In total, 46 research groups participated in the shared task and 7 experimental runs were submitted for evaluation. We used BLEU scores for assessment of the translations.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } This paper presents an overview of the shared task on machine translation of Dravidian languages. We presented the shared task results at the EACL 2021 workshop on Speech and Language Technologies for Dravidian Languages. This paper describes the datasets used, the methodology used for the evaluation of participants, and the experiments' overall results. As a part of this shared task, we organized four sub-tasks corresponding to machine translation of the following language pairs: English to Tamil, English to Malayalam, English to Telugu and Tamil to Telugu which are available at https://competitions.codalab.org/competitions/27650. We provided the participants with training and development datasets to perform experiments, and the results were evaluated on unseen test data. In total, 46 research groups participated in the shared task and 7 experimental runs were submitted for evaluation. We used BLEU scores for assessment of the translations. |
Bharathi Raja Chakravarthi; Ruba Priyadharshini; Navya Jose; Anand Kumar M; Thomas Mandl; Prasanna Kumar Kumaresan; Rahul Ponnusamy; Hariharan R L; John P McCrae; Elizabeth Sherly Findings of the Shared Task on Offensive Language Identification in Tamil, Malayalam, and Kannada Inproceedings Proceedings of the First Workshop on Speech and Language Technologies for Dravidian Languages, pp. 133–145, Association for Computational Linguistics, Kyiv, 2021. @inproceedings{chakravarthi-etal-2021-findings-shared, title = {Findings of the Shared Task on Offensive Language Identification in Tamil, Malayalam, and Kannada}, author = {Bharathi Raja Chakravarthi and Ruba Priyadharshini and Navya Jose and Anand Kumar M and Thomas Mandl and Prasanna Kumar Kumaresan and Rahul Ponnusamy and Hariharan R L and John P McCrae and Elizabeth Sherly}, url = {https://www.aclweb.org/anthology/2021.dravidianlangtech-1.17}, year = {2021}, date = {2021-01-01}, booktitle = {Proceedings of the First Workshop on Speech and Language Technologies for Dravidian Languages}, pages = {133--145}, publisher = {Association for Computational Linguistics}, address = {Kyiv}, abstract = {Detecting offensive language in social media in local languages is critical for moderating user-generated content. Thus, the field of offensive language identification in under-resourced Tamil, Malayalam and Kannada languages are essential. As the user-generated content is more code-mixed and not well studied for under-resourced languages, it is imperative to create resources and conduct benchmarking studies to encourage research in under-resourced Dravidian languages. We created a shared task on offensive language detection in Dravidian languages. We summarize here the dataset for this challenge which are openly available at https://competitions.codalab.org/competitions/27654, and present an overview of the methods and the results of the competing systems.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Detecting offensive language in social media in local languages is critical for moderating user-generated content. Thus, the field of offensive language identification in under-resourced Tamil, Malayalam and Kannada languages are essential. As the user-generated content is more code-mixed and not well studied for under-resourced languages, it is imperative to create resources and conduct benchmarking studies to encourage research in under-resourced Dravidian languages. We created a shared task on offensive language detection in Dravidian languages. We summarize here the dataset for this challenge which are openly available at https://competitions.codalab.org/competitions/27654, and present an overview of the methods and the results of the competing systems. |
2020 |
Koustava Goswami; Priya Rani; Bharathi Raja Chakravarthi; Theodorus Fransen; John P McCrae Proceedings of the Fourteenth Workshop on Semantic Evaluation, pp. 968–974, International Committee for Computational Linguistics, Barcelona (online), 2020. @inproceedings{goswami-etal-2020-uld, title = {[email protected] at SemEval-2020 Task 9: Generative Morphemes with an Attention Model for Sentiment Analysis in Code-Mixed Text}, author = {Koustava Goswami and Priya Rani and Bharathi Raja Chakravarthi and Theodorus Fransen and John P McCrae}, url = {https://www.aclweb.org/anthology/2020.semeval-1.125}, year = {2020}, date = {2020-12-01}, booktitle = {Proceedings of the Fourteenth Workshop on Semantic Evaluation}, pages = {968--974}, publisher = {International Committee for Computational Linguistics}, address = {Barcelona (online)}, abstract = {Code mixing is a common phenomena in multilingual societies where people switch from one language to another for various reasons. Recent advances in public communication over different social media sites have led to an increase in the frequency of code-mixed usage in written language. In this paper, we present the Generative Morphemes with Attention (GenMA) Model sentiment analysis system contributed to SemEval 2020 Task 9 SentiMix. The system aims to predict the sentiments of the given English-Hindi code-mixed tweets without using word-level language tags instead inferring this automatically using a morphological model. The system is based on a novel deep neural network (DNN) architecture, which has outperformed the baseline F1-score on the test data-set as well as the validation data-set. Our results can be found under the user name ``koustava'' on the ``Sentimix Hindi English'' page.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Code mixing is a common phenomena in multilingual societies where people switch from one language to another for various reasons. Recent advances in public communication over different social media sites have led to an increase in the frequency of code-mixed usage in written language. In this paper, we present the Generative Morphemes with Attention (GenMA) Model sentiment analysis system contributed to SemEval 2020 Task 9 SentiMix. The system aims to predict the sentiments of the given English-Hindi code-mixed tweets without using word-level language tags instead inferring this automatically using a morphological model. The system is based on a novel deep neural network (DNN) architecture, which has outperformed the baseline F1-score on the test data-set as well as the validation data-set. Our results can be found under the user name ``koustava'' on the ``Sentimix Hindi English'' page. |
Koustava Goswami; Rajdeep Sarkar; Bharathi Raja Chakravarthi; Theodorus Fransen; John P McCrae Unsupervised Deep Language and Dialect Identification for Short Texts Inproceedings Proceedings of the 28th International Conference on Computational Linguistics, pp. 1606–1617, International Committee on Computational Linguistics, Barcelona, Spain (Online), 2020. @inproceedings{goswami-etal-2020-unsupervised, title = {Unsupervised Deep Language and Dialect Identification for Short Texts}, author = {Koustava Goswami and Rajdeep Sarkar and Bharathi Raja Chakravarthi and Theodorus Fransen and John P McCrae}, url = {https://www.aclweb.org/anthology/2020.coling-main.141}, doi = {10.18653/v1/2020.coling-main.141}, year = {2020}, date = {2020-12-01}, booktitle = {Proceedings of the 28th International Conference on Computational Linguistics}, pages = {1606--1617}, publisher = {International Committee on Computational Linguistics}, address = {Barcelona, Spain (Online)}, abstract = {Automatic Language Identification (LI) or Dialect Identification (DI) of short texts of closely related languages or dialects, is one of the primary steps in many natural language processing pipelines. Language identification is considered a solved task in many cases; however, in the case of very closely related languages, or in an unsupervised scenario (where the languages are not known in advance), performance is still poor. In this paper, we propose the Unsupervised Deep Language and Dialect Identification (UDLDI)method, whichcan simultaneously learn sentence embeddings and cluster assignments from short texts. The UDLDI modelunderstands the sentence constructions of languages by applying attention to character relations which helps to optimize the clustering of languages. We have performed our experiments on three short-text datasets for different language families, each consisting of closely related languages or dialects, with very minimal training sets. Our experimental evaluations on these datasets have shown significant improvement over state-of-the-artunsupervised methods and our model has outperformed state-of-the-art LI and DI systems in supervised settings.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Automatic Language Identification (LI) or Dialect Identification (DI) of short texts of closely related languages or dialects, is one of the primary steps in many natural language processing pipelines. Language identification is considered a solved task in many cases; however, in the case of very closely related languages, or in an unsupervised scenario (where the languages are not known in advance), performance is still poor. In this paper, we propose the Unsupervised Deep Language and Dialect Identification (UDLDI)method, whichcan simultaneously learn sentence embeddings and cluster assignments from short texts. The UDLDI modelunderstands the sentence constructions of languages by applying attention to character relations which helps to optimize the clustering of languages. We have performed our experiments on three short-text datasets for different language families, each consisting of closely related languages or dialects, with very minimal training sets. Our experimental evaluations on these datasets have shown significant improvement over state-of-the-artunsupervised methods and our model has outperformed state-of-the-art LI and DI systems in supervised settings. |
Adeep Hande; Ruba Priyadharshini; Bharathi Raja Chakravarthi KanCMD: Kannada CodeMixed Dataset for Sentiment Analysis and Offensive Language Detection Inproceedings Proceedings of the Third Workshop on Computational Modeling of People's Opinions, Personality, and Emotion's in Social Media, pp. 54–63, Association for Computational Linguistics, Barcelona, Spain (Online), 2020. @inproceedings{hande-etal-2020-kancmd, title = {KanCMD: Kannada CodeMixed Dataset for Sentiment Analysis and Offensive Language Detection}, author = {Adeep Hande and Ruba Priyadharshini and Bharathi Raja Chakravarthi}, url = {https://www.aclweb.org/anthology/2020.peoples-1.6}, year = {2020}, date = {2020-12-01}, booktitle = {Proceedings of the Third Workshop on Computational Modeling of People's Opinions, Personality, and Emotion's in Social Media}, pages = {54--63}, publisher = {Association for Computational Linguistics}, address = {Barcelona, Spain (Online)}, abstract = {We introduce Kannada CodeMixed Dataset (KanCMD), a multi-task learning dataset for sentiment analysis and offensive language identification. The KanCMD dataset highlights two real-world issues from the social media text. First, it contains actual comments in code mixed text posted by users on YouTube social media, rather than in monolingual text from the textbook. Second, it has been annotated for two tasks, namely sentiment analysis and offensive language detection for under-resourced Kannada language. Hence, KanCMD is meant to stimulate research in under-resourced Kannada language on real-world code-mixed social media text and multi-task learning. KanCMD was obtained by crawling the YouTube, and a minimum of three annotators annotates each comment. We release KanCMD 7,671 comments for multitask learning research purpose.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } We introduce Kannada CodeMixed Dataset (KanCMD), a multi-task learning dataset for sentiment analysis and offensive language identification. The KanCMD dataset highlights two real-world issues from the social media text. First, it contains actual comments in code mixed text posted by users on YouTube social media, rather than in monolingual text from the textbook. Second, it has been annotated for two tasks, namely sentiment analysis and offensive language detection for under-resourced Kannada language. Hence, KanCMD is meant to stimulate research in under-resourced Kannada language on real-world code-mixed social media text and multi-task learning. KanCMD was obtained by crawling the YouTube, and a minimum of three annotators annotates each comment. We release KanCMD 7,671 comments for multitask learning research purpose. |
Bharathi Raja Chakravarthi HopeEDI: A Multilingual Hope Speech Detection Dataset for Equality, Diversity, and Inclusion Inproceedings Proceedings of the Third Workshop on Computational Modeling of People's Opinions, Personality, and Emotion's in Social Media, pp. 41–53, Association for Computational Linguistics, Barcelona, Spain (Online), 2020. @inproceedings{chakravarthi-2020-hopeedi, title = {HopeEDI: A Multilingual Hope Speech Detection Dataset for Equality, Diversity, and Inclusion}, author = {Bharathi Raja Chakravarthi}, url = {https://www.aclweb.org/anthology/2020.peoples-1.5}, year = {2020}, date = {2020-12-01}, booktitle = {Proceedings of the Third Workshop on Computational Modeling of People's Opinions, Personality, and Emotion's in Social Media}, pages = {41--53}, publisher = {Association for Computational Linguistics}, address = {Barcelona, Spain (Online)}, abstract = {Over the past few years, systems have been developed to control online content and eliminate abusive, offensive or hate speech content. However, people in power sometimes misuse this form of censorship to obstruct the democratic right of freedom of speech. Therefore, it is imperative that research should take a positive reinforcement approach towards online content that is encouraging, positive and supportive contents. Until now, most studies have focused on solving this problem of negativity in the English language, though the problem is much more than just harmful content. Furthermore, it is multilingual as well. Thus, we have constructed a Hope Speech dataset for Equality, Diversity and Inclusion (HopeEDI) containing user-generated comments from the social media platform YouTube with 28,451, 20,198 and 10,705 comments in English, Tamil and Malayalam, respectively, manually labelled as containing hope speech or not. To our knowledge, this is the first research of its kind to annotate hope speech for equality, diversity and inclusion in a multilingual setting. We determined that the inter-annotator agreement of our dataset using Krippendorff's alpha. Further, we created several baselines to benchmark the resulting dataset and the results have been expressed using precision, recall and F1-score. The dataset is publicly available for the research community. We hope that this resource will spur further research on encouraging inclusive and responsive speech that reinforces positiveness.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Over the past few years, systems have been developed to control online content and eliminate abusive, offensive or hate speech content. However, people in power sometimes misuse this form of censorship to obstruct the democratic right of freedom of speech. Therefore, it is imperative that research should take a positive reinforcement approach towards online content that is encouraging, positive and supportive contents. Until now, most studies have focused on solving this problem of negativity in the English language, though the problem is much more than just harmful content. Furthermore, it is multilingual as well. Thus, we have constructed a Hope Speech dataset for Equality, Diversity and Inclusion (HopeEDI) containing user-generated comments from the social media platform YouTube with 28,451, 20,198 and 10,705 comments in English, Tamil and Malayalam, respectively, manually labelled as containing hope speech or not. To our knowledge, this is the first research of its kind to annotate hope speech for equality, diversity and inclusion in a multilingual setting. We determined that the inter-annotator agreement of our dataset using Krippendorff's alpha. Further, we created several baselines to benchmark the resulting dataset and the results have been expressed using precision, recall and F1-score. The dataset is publicly available for the research community. We hope that this resource will spur further research on encouraging inclusive and responsive speech that reinforces positiveness. |
Theodorus Fransen Rekenen met taal: computationele taalkunde en historisch Iers Journal Article Kelten: Mededelingen van de Stichting A. G. van Hamel voor Keltische Studies, 84 , 2020. @article{fransen-kelten-2020, title = {Rekenen met taal: computationele taalkunde en historisch Iers}, author = {Theodorus Fransen}, url = {https://kelten.vanhamel.nl/k84-2020-fransen-oudiers-computationele-taalkunde-morfologie-deep-learning}, year = {2020}, date = {2020-08-17}, journal = {Kelten: Mededelingen van de Stichting A. G. van Hamel voor Keltische Studies}, volume = {84}, abstract = {Language and mathematics meet in the dynamic field of computational linguistics. Using the latest deep learning techniques and transferring results from better resourced languages, Theodorus Fransen of the Cardamom project in Galway hopes to develop an algorithm capable of analysing the various forms of the Irish verb across time and space, adding to the linguistic tools available to both students and scholars of historical Irish.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Language and mathematics meet in the dynamic field of computational linguistics. Using the latest deep learning techniques and transferring results from better resourced languages, Theodorus Fransen of the Cardamom project in Galway hopes to develop an algorithm capable of analysing the various forms of the Irish verb across time and space, adding to the linguistic tools available to both students and scholars of historical Irish. |
Bharathi Raja Chakravarthi Leveraging orthographic information to improve machine translation of under-resourced languages PhD Thesis National University of Ireland Galway, 2020. @phdthesis{chakravarthi-phd-2020, title = {Leveraging orthographic information to improve machine translation of under-resourced languages}, author = {Bharathi Raja Chakravarthi}, doi = {http://hdl.handle.net/10379/16100}, year = {2020}, date = {2020-07-27}, school = {National University of Ireland Galway}, abstract = {This thesis describes our improvement of word sense translation for under-resourced languages utilizing orthographic information with a particular focus on creating resources using machine translation. The first target of this thesis is cleaning the noisy corpus in the form of code-mixed content at word-level based on orthographic information to improve machine translation quality. Our results indicate that the proposed removing of code-mixed text based on orthography results in improvement for Dravidian languages. We then turn our interest to the usage of training data from closely-related languages. While languages within the same language family share many properties, many under-resourced languages are written in their own native script, which makes taking advantage of these language similarities difficult. We propose to alleviate the problem of different scripts by transcribing the native script into a common representation such as the Latin script or the International Phonetic Alphabet (IPA). We also show that our method could aid the creation or improvement of wordnets for under-resourced languages using machine translation. Further, we investigate bilingual lexicon induction using pre-trained monolingual word embeddings and orthographic information. We use existing resources such as IndoWordNet entries as a seed dictionary and test set for the under-resourced Dravidian languages. To take advantage of orthographic information, we propose to bring the related languages into a single script before creating word embeddings, and use the longest common subsequence to take advantage of cognate information. Our methods for under-resourced word sense translation of Dravidian languages outperformed state-of-the art systems in terms of both automatic and manual evaluation.}, keywords = {}, pubstate = {published}, tppubtype = {phdthesis} } This thesis describes our improvement of word sense translation for under-resourced languages utilizing orthographic information with a particular focus on creating resources using machine translation. The first target of this thesis is cleaning the noisy corpus in the form of code-mixed content at word-level based on orthographic information to improve machine translation quality. Our results indicate that the proposed removing of code-mixed text based on orthography results in improvement for Dravidian languages. We then turn our interest to the usage of training data from closely-related languages. While languages within the same language family share many properties, many under-resourced languages are written in their own native script, which makes taking advantage of these language similarities difficult. We propose to alleviate the problem of different scripts by transcribing the native script into a common representation such as the Latin script or the International Phonetic Alphabet (IPA). We also show that our method could aid the creation or improvement of wordnets for under-resourced languages using machine translation. Further, we investigate bilingual lexicon induction using pre-trained monolingual word embeddings and orthographic information. We use existing resources such as IndoWordNet entries as a seed dictionary and test set for the under-resourced Dravidian languages. To take advantage of orthographic information, we propose to bring the related languages into a single script before creating word embeddings, and use the longest common subsequence to take advantage of cognate information. Our methods for under-resourced word sense translation of Dravidian languages outperformed state-of-the art systems in terms of both automatic and manual evaluation. |
Francis Bond; Luis Morgado da Costa; Michael Wayne Goodman; John Philip McCrae; Ahti Lohk Some Issues with Building a Multilingual Wordnet Inproceedings Proceedings of the 12th Language Resources and Evaluation Conference, pp. 3189–3197, European Language Resources Association, Marseille, France, 2020, ISBN: 979-10-95546-34-4. @inproceedings{bond-etal-2020-issues, title = {Some Issues with Building a Multilingual Wordnet}, author = {Francis Bond and Luis Morgado da Costa and Michael Wayne Goodman and John Philip McCrae and Ahti Lohk}, url = {https://www.aclweb.org/anthology/2020.lrec-1.390}, isbn = {979-10-95546-34-4}, year = {2020}, date = {2020-05-01}, booktitle = {Proceedings of the 12th Language Resources and Evaluation Conference}, pages = {3189--3197}, publisher = {European Language Resources Association}, address = {Marseille, France}, abstract = {In this paper we discuss the experience of bringing together over 40 different wordnets. We introduce some extensions to the GWA wordnet LMF format proposed in Vossen et al. (2016) and look at how this new information can be displayed. Notable extensions include: confidence, corpus frequency, orthographic variants, lexicalized and non-lexicalized synsets and lemmas, new parts of speech, and more. Many of these extensions already exist in multiple wordnets -- the challenge was to find a compatible representation. To this end, we introduce a new version of the Open Multilingual Wordnet (Bond and Foster, 2013), that integrates a new set of tools that tests the extensions introduced by this new format, while also ensuring the integrity of the Collaborative Interlingual Index (CILI: Bond et al., 2016), avoiding the same new concept to be introduced through multiple projects.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } In this paper we discuss the experience of bringing together over 40 different wordnets. We introduce some extensions to the GWA wordnet LMF format proposed in Vossen et al. (2016) and look at how this new information can be displayed. Notable extensions include: confidence, corpus frequency, orthographic variants, lexicalized and non-lexicalized synsets and lemmas, new parts of speech, and more. Many of these extensions already exist in multiple wordnets -- the challenge was to find a compatible representation. To this end, we introduce a new version of the Open Multilingual Wordnet (Bond and Foster, 2013), that integrates a new set of tools that tests the extensions introduced by this new format, while also ensuring the integrity of the Collaborative Interlingual Index (CILI: Bond et al., 2016), avoiding the same new concept to be introduced through multiple projects. |
Bharathi Raja Chakravarthi; Navya Jose; Shardul Suryawanshi; Elizabeth Sherly; John Philip McCrae A Sentiment Analysis Dataset for Code-Mixed Malayalam-English Inproceedings Proceedings of the 1st Joint Workshop on Spoken Language Technologies for Under-resourced languages (SLTU) and Collaboration and Computing for Under-Resourced Languages (CCURL), pp. 177–184, European Language Resources association, Marseille, France, 2020, ISBN: 979-10-95546-35-1. @inproceedings{chakravarthi-etal-2020-sentiment, title = {A Sentiment Analysis Dataset for Code-Mixed Malayalam-English}, author = {Bharathi Raja Chakravarthi and Navya Jose and Shardul Suryawanshi and Elizabeth Sherly and John Philip McCrae}, url = {https://www.aclweb.org/anthology/2020.sltu-1.25}, isbn = {979-10-95546-35-1}, year = {2020}, date = {2020-05-01}, booktitle = {Proceedings of the 1st Joint Workshop on Spoken Language Technologies for Under-resourced languages (SLTU) and Collaboration and Computing for Under-Resourced Languages (CCURL)}, pages = {177--184}, publisher = {European Language Resources association}, address = {Marseille, France}, abstract = {There is an increasing demand for sentiment analysis of text from social media which are mostly code-mixed. Systems trained on monolingual data fail for code-mixed data due to the complexity of mixing at different levels of the text. However, very few resources are available for code-mixed data to create models specific for this data. Although much research in multilingual and cross-lingual sentiment analysis has used semi-supervised or unsupervised methods, supervised methods still performs better. Only a few datasets for popular languages such as English-Spanish, English-Hindi, and English-Chinese are available. There are no resources available for Malayalam-English code-mixed data. This paper presents a new gold standard corpus for sentiment analysis of code-mixed text in Malayalam-English annotated by voluntary annotators. This gold standard corpus obtained a Krippendorff's alpha above 0.8 for the dataset. We use this new corpus to provide the benchmark for sentiment analysis in Malayalam-English code-mixed texts.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } There is an increasing demand for sentiment analysis of text from social media which are mostly code-mixed. Systems trained on monolingual data fail for code-mixed data due to the complexity of mixing at different levels of the text. However, very few resources are available for code-mixed data to create models specific for this data. Although much research in multilingual and cross-lingual sentiment analysis has used semi-supervised or unsupervised methods, supervised methods still performs better. Only a few datasets for popular languages such as English-Spanish, English-Hindi, and English-Chinese are available. There are no resources available for Malayalam-English code-mixed data. This paper presents a new gold standard corpus for sentiment analysis of code-mixed text in Malayalam-English annotated by voluntary annotators. This gold standard corpus obtained a Krippendorff's alpha above 0.8 for the dataset. We use this new corpus to provide the benchmark for sentiment analysis in Malayalam-English code-mixed texts. |
Bharathi Raja Chakravarthi; Vigneshwaran Muralidaran; Ruba Priyadharshini; John Philip McCrae Corpus Creation for Sentiment Analysis in Code-Mixed Tamil-English Text Inproceedings Proceedings of the 1st Joint Workshop on Spoken Language Technologies for Under-resourced languages (SLTU) and Collaboration and Computing for Under-Resourced Languages (CCURL), pp. 202–210, European Language Resources association, Marseille, France, 2020, ISBN: 979-10-95546-35-1. @inproceedings{chakravarthi-etal-2020-corpus, title = {Corpus Creation for Sentiment Analysis in Code-Mixed Tamil-English Text}, author = {Bharathi Raja Chakravarthi and Vigneshwaran Muralidaran and Ruba Priyadharshini and John Philip McCrae}, url = {https://www.aclweb.org/anthology/2020.sltu-1.28}, isbn = {979-10-95546-35-1}, year = {2020}, date = {2020-05-01}, booktitle = {Proceedings of the 1st Joint Workshop on Spoken Language Technologies for Under-resourced languages (SLTU) and Collaboration and Computing for Under-Resourced Languages (CCURL)}, pages = {202--210}, publisher = {European Language Resources association}, address = {Marseille, France}, abstract = {Understanding the sentiment of a comment from a video or an image is an essential task in many applications. Sentiment analysis of a text can be useful for various decision-making processes. One such application is to analyse the popular sentiments of videos on social media based on viewer comments. However, comments from social media do not follow strict rules of grammar, and they contain mixing of more than one language, often written in non-native scripts. Non-availability of annotated code-mixed data for a low-resourced language like Tamil also adds difficulty to this problem. To overcome this, we created a gold standard Tamil-English code-switched, sentiment-annotated corpus containing 15,744 comment posts from YouTube. In this paper, we describe the process of creating the corpus and assigning polarities. We present inter-annotator agreement and show the results of sentiment analysis trained on this corpus as a benchmark.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Understanding the sentiment of a comment from a video or an image is an essential task in many applications. Sentiment analysis of a text can be useful for various decision-making processes. One such application is to analyse the popular sentiments of videos on social media based on viewer comments. However, comments from social media do not follow strict rules of grammar, and they contain mixing of more than one language, often written in non-native scripts. Non-availability of annotated code-mixed data for a low-resourced language like Tamil also adds difficulty to this problem. To overcome this, we created a gold standard Tamil-English code-switched, sentiment-annotated corpus containing 15,744 comment posts from YouTube. In this paper, we describe the process of creating the corpus and assigning polarities. We present inter-annotator agreement and show the results of sentiment analysis trained on this corpus as a benchmark. |
Priya Rani; Shardul Suryawanshi; Koustava Goswami; Bharathi Raja Chakravarthi; Theodorus Fransen; John Philip McCrae A Comparative Study of Different State-of-the-Art Hate Speech Detection Methods in Hindi-English Code-Mixed Data Inproceedings Proceedings of the Second Workshop on Trolling, Aggression and Cyberbullying, pp. 42–48, European Language Resources Association (ELRA), Marseille, France, 2020. @inproceedings{rani-etal-2020-comparative, title = {A Comparative Study of Different State-of-the-Art Hate Speech Detection Methods in Hindi-English Code-Mixed Data}, author = {Priya Rani and Shardul Suryawanshi and Koustava Goswami and Bharathi Raja Chakravarthi and Theodorus Fransen and John Philip McCrae}, url = {https://www.aclweb.org/anthology/2020.trac-1.7}, year = {2020}, date = {2020-05-01}, booktitle = {Proceedings of the Second Workshop on Trolling, Aggression and Cyberbullying}, pages = {42--48}, publisher = {European Language Resources Association (ELRA)}, address = {Marseille, France}, abstract = {Hate speech detection in social media communication has become one of the primary concerns to avoid conflicts and curb undesired activities. In an environment where multilingual speakers switch among multiple languages, hate speech detection becomes a challenging task using methods that are designed for monolingual corpora. In our work, we attempt to analyze, detect and provide a comparative study of hate speech in a code-mixed social media text. We also provide a Hindi-English code-mixed data set consisting of Facebook and Twitter posts and comments. Our experiments show that deep learning models trained on this code-mixed corpus perform better.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Hate speech detection in social media communication has become one of the primary concerns to avoid conflicts and curb undesired activities. In an environment where multilingual speakers switch among multiple languages, hate speech detection becomes a challenging task using methods that are designed for monolingual corpora. In our work, we attempt to analyze, detect and provide a comparative study of hate speech in a code-mixed social media text. We also provide a Hindi-English code-mixed data set consisting of Facebook and Twitter posts and comments. Our experiments show that deep learning models trained on this code-mixed corpus perform better. |
Bharathi Raja Chakravarthi; Navaneethan Rajasekaran; Mihael Arcan; Kevin McGuinness; Noel E. O'Connor; John P McCrae Bilingual Lexicon Induction across Orthographically-distinct Under-Resourced Dravidian Languages Inproceedings Proceedings of the 7th Workshop on NLP for Similar Languages, Varieties and Dialects, pp. 57–69, International Committee on Computational Linguistics (ICCL), Barcelona, Spain (Online), 2020. @inproceedings{chakravarthi-etal-2020-bilingual, title = {Bilingual Lexicon Induction across Orthographically-distinct Under-Resourced Dravidian Languages}, author = {Bharathi Raja Chakravarthi and Navaneethan Rajasekaran and Mihael Arcan and Kevin McGuinness and Noel E. O{'}Connor and John P McCrae}, url = {https://www.aclweb.org/anthology/2020.vardial-1.6}, year = {2020}, date = {2020-01-01}, booktitle = {Proceedings of the 7th Workshop on NLP for Similar Languages, Varieties and Dialects}, pages = {57--69}, publisher = {International Committee on Computational Linguistics (ICCL)}, address = {Barcelona, Spain (Online)}, abstract = {Bilingual lexicons are a vital tool for under-resourced languages and recent state-of-the-art approaches to this leverage pretrained monolingual word embeddings using supervised or semi-supervised approaches. However, these approaches require cross-lingual information such as seed dictionaries to train the model and find a linear transformation between the word embedding spaces. Especially in the case of low-resourced languages, seed dictionaries are not readily available, and as such, these methods produce extremely weak results on these languages. In this work, we focus on the Dravidian languages, namely Tamil, Telugu, Kannada, and Malayalam, which are even more challenging as they are written in unique scripts. To take advantage of orthographic information and cognates in these languages, we bring the related languages into a single script. Previous approaches have used linguistically sub-optimal measures such as the Levenshtein edit distance to detect cognates, whereby we demonstrate that the longest common sub-sequence is linguistically more sound and improves the performance of bilingual lexicon induction. We show that our approach can increase the accuracy of bilingual lexicon induction methods on these languages many times, making bilingual lexicon induction approaches feasible for such under-resourced languages.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Bilingual lexicons are a vital tool for under-resourced languages and recent state-of-the-art approaches to this leverage pretrained monolingual word embeddings using supervised or semi-supervised approaches. However, these approaches require cross-lingual information such as seed dictionaries to train the model and find a linear transformation between the word embedding spaces. Especially in the case of low-resourced languages, seed dictionaries are not readily available, and as such, these methods produce extremely weak results on these languages. In this work, we focus on the Dravidian languages, namely Tamil, Telugu, Kannada, and Malayalam, which are even more challenging as they are written in unique scripts. To take advantage of orthographic information and cognates in these languages, we bring the related languages into a single script. Previous approaches have used linguistically sub-optimal measures such as the Levenshtein edit distance to detect cognates, whereby we demonstrate that the longest common sub-sequence is linguistically more sound and improves the performance of bilingual lexicon induction. We show that our approach can increase the accuracy of bilingual lexicon induction methods on these languages many times, making bilingual lexicon induction approaches feasible for such under-resourced languages. |
2019 |
John P McCrae; Theodorus Fransen Cardamom: Comparative Deep Models for Minority and Historical Languages Inproceedings Proceedings of the 1st International Conference on Language Technologies for All, pp. 276–279, European Language Resources Association (ELRA), Paris, France, 2019. @inproceedings{mccrae-fransen:2019:LT4ALL, title = {Cardamom: Comparative Deep Models for Minority and Historical Languages}, author = {John P McCrae and Theodorus Fransen}, url = {https://lt4all.elra.info/proceedings/lt4all2019/index.html}, year = {2019}, date = {2019-12-01}, booktitle = {Proceedings of the 1st International Conference on Language Technologies for All}, pages = {276--279}, publisher = {European Language Resources Association (ELRA)}, address = {Paris, France}, abstract = {This paper gives an overview of the Cardamom project, which aims to close the resource gap for minority and under-resourced languages by means of deep-learning-based natural language processing (NLP) and exploiting similarities of closely-related languages. The project further extends this idea to historical languages, which can be considered as closely related to their modern form, and as such aims to provide NLP through both space and time for languages that have been ignored by current approaches.}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } This paper gives an overview of the Cardamom project, which aims to close the resource gap for minority and under-resourced languages by means of deep-learning-based natural language processing (NLP) and exploiting similarities of closely-related languages. The project further extends this idea to historical languages, which can be considered as closely related to their modern form, and as such aims to provide NLP through both space and time for languages that have been ignored by current approaches. |
Publications
2021 |
Findings of the Shared Task on Troll Meme Classification in Tamil Inproceedings Proceedings of the First Workshop on Speech and Language Technologies for Dravidian Languages, pp. 126–132, Association for Computational Linguistics, Kyiv, 2021. |
Findings of the Shared Task on Machine Translation in Dravidian languages Inproceedings Proceedings of the First Workshop on Speech and Language Technologies for Dravidian Languages, pp. 119–125, Association for Computational Linguistics, Kyiv, 2021. |
Findings of the Shared Task on Offensive Language Identification in Tamil, Malayalam, and Kannada Inproceedings Proceedings of the First Workshop on Speech and Language Technologies for Dravidian Languages, pp. 133–145, Association for Computational Linguistics, Kyiv, 2021. |
2020 |
Proceedings of the Fourteenth Workshop on Semantic Evaluation, pp. 968–974, International Committee for Computational Linguistics, Barcelona (online), 2020. |
Unsupervised Deep Language and Dialect Identification for Short Texts Inproceedings Proceedings of the 28th International Conference on Computational Linguistics, pp. 1606–1617, International Committee on Computational Linguistics, Barcelona, Spain (Online), 2020. |
KanCMD: Kannada CodeMixed Dataset for Sentiment Analysis and Offensive Language Detection Inproceedings Proceedings of the Third Workshop on Computational Modeling of People's Opinions, Personality, and Emotion's in Social Media, pp. 54–63, Association for Computational Linguistics, Barcelona, Spain (Online), 2020. |
HopeEDI: A Multilingual Hope Speech Detection Dataset for Equality, Diversity, and Inclusion Inproceedings Proceedings of the Third Workshop on Computational Modeling of People's Opinions, Personality, and Emotion's in Social Media, pp. 41–53, Association for Computational Linguistics, Barcelona, Spain (Online), 2020. |
Rekenen met taal: computationele taalkunde en historisch Iers Journal Article Kelten: Mededelingen van de Stichting A. G. van Hamel voor Keltische Studies, 84 , 2020. |
Leveraging orthographic information to improve machine translation of under-resourced languages PhD Thesis National University of Ireland Galway, 2020. |
Some Issues with Building a Multilingual Wordnet Inproceedings Proceedings of the 12th Language Resources and Evaluation Conference, pp. 3189–3197, European Language Resources Association, Marseille, France, 2020, ISBN: 979-10-95546-34-4. |
A Sentiment Analysis Dataset for Code-Mixed Malayalam-English Inproceedings Proceedings of the 1st Joint Workshop on Spoken Language Technologies for Under-resourced languages (SLTU) and Collaboration and Computing for Under-Resourced Languages (CCURL), pp. 177–184, European Language Resources association, Marseille, France, 2020, ISBN: 979-10-95546-35-1. |
Corpus Creation for Sentiment Analysis in Code-Mixed Tamil-English Text Inproceedings Proceedings of the 1st Joint Workshop on Spoken Language Technologies for Under-resourced languages (SLTU) and Collaboration and Computing for Under-Resourced Languages (CCURL), pp. 202–210, European Language Resources association, Marseille, France, 2020, ISBN: 979-10-95546-35-1. |
A Comparative Study of Different State-of-the-Art Hate Speech Detection Methods in Hindi-English Code-Mixed Data Inproceedings Proceedings of the Second Workshop on Trolling, Aggression and Cyberbullying, pp. 42–48, European Language Resources Association (ELRA), Marseille, France, 2020. |
Bilingual Lexicon Induction across Orthographically-distinct Under-Resourced Dravidian Languages Inproceedings Proceedings of the 7th Workshop on NLP for Similar Languages, Varieties and Dialects, pp. 57–69, International Committee on Computational Linguistics (ICCL), Barcelona, Spain (Online), 2020. |
2019 |
Cardamom: Comparative Deep Models for Minority and Historical Languages Inproceedings Proceedings of the 1st International Conference on Language Technologies for All, pp. 276–279, European Language Resources Association (ELRA), Paris, France, 2019. |