Sentiment Analysis (SA) employing code-mixed data from social media helps in getting insights to the data and decision making for various applications. One such application is to analyze users’ emotions from comments of videos on YouTube. Social media comments do not adhere to the grammatical norms of any language and they often comprise a mix of languages and scripts. The lack of annotated code-mixed data for SA in a low-resource language like Tulu makes the SA a challenging task. To address the lack of annotated code-mixed Tulu data for SA, a gold standard trlingual code-mixed Tulu annotated corpus of 7,171 YouTube comments is created. Further, Machine Learning (ML) algorithms are employed as baseline models to evaluate the developed dataset and the performance of the ML algorithms are found to be encouraging.
Thirumurai, also known as Panniru Thirumurai, is a collection of Tamil Shaivite poems dating back to the Hindu revival period between the 6th and the 10th century. These poems are par excellence, in both literary and musical terms. They have been composed based on the ancient, now non-existent Tamil Pann system and can be set to music. We present a large dataset containing all the Thirumurai poems and also attempt to classify the Pann and author of each poem using transformer based architectures. Our work is the first of its kind in dealing with ancient Tamil text datasets, which are severely under-resourced. We explore several Deep Learning-based techniques for solving this challenge effectively and provide essential insights into the problem and how to address it.
Medical dialogue systems have the potential to assist doctors in expanding access to medical care, improving the quality of patient experiences, and lowering medical expenses. The computational methods are still in their early stages and are not ready for widespread application despite their great potential. Existing transformer-based language models have shown promising results but lack domain-specific knowledge. However, to diagnose like doctors, an automatic medical diagnosis necessitates more stringent requirements for the rationality of the dialogue in the context of relevant knowledge. In this study, we propose a new method that addresses the challenges of medical dialogue generation by incorporating medical knowledge into transformer-based language models. We present a method that leverages an external medical knowledge graph and injects triples as domain knowledge into the utterances. Automatic and human evaluation on a publicly available dataset demonstrates that incorporating medical knowledge outperforms several state-of-the-art baseline methods.
Social media is an idea created to make theworld smaller and more connected. Recently,it has become a hub of fake news and sexistmemes that target women. Social Media shouldensure proper women’s safety and equality. Filteringsuch information from social media is ofparamount importance to achieving this goal.In this paper, we describe the system developedby our team for SemEval-2022 Task 5: MultimediaAutomatic Misogyny Identification. Wepropose a multimodal training methodologythat achieves good performance on both thesubtasks, ranking 4th for Subtask A (0.718macro F1-score) and 9th for Subtask B (0.695macro F1-score) while exceeding the baselineresults by good margins.
Automatic Term Extraction (ATE) is one of the core problems in natural language processing and forms a key component of text mining pipelines of domain specific corpora. Complex low-level tasks such as machine translation and summarization for domain specific texts necessitate the use of term extraction systems. However, the development of these systems requires the use of large annotated datasets and thus there has been little progress made on this front for under-resourced languages. As a part of ongoing research, we present a dataset for term extraction from Hindi texts in this paper. To the best of our knowledge, this is the first dataset that provides term annotated documents for Hindi. Furthermore, we have evaluated this dataset on statistical term extraction methods and the results obtained indicate the problems associated with development of term extractors for under-resourced languages.
With the rise of social media and internet, thereis a necessity to provide an inclusive space andprevent the abusive topics against any gender,race or community. This paper describes thesystem submitted to the ACL-2022 shared taskon fine-grained abuse detection in Tamil. In ourapproach we transliterated code-mixed datasetas an augmentation technique to increase thesize of the data. Using this method we wereable to rank 3rd on the task with a 0.290 macroaverage F1 score and a 0.590 weighted F1score
This paper investigates the effectiveness of sentence-level transformers for zero-shot offensive span identification on a code-mixed Tamil dataset. More specifically, we evaluate rationale extraction methods of Local Interpretable Model Agnostic Explanations (LIME) (CITATION) and Integrated Gradients (IG) (CITATION) for adapting transformer based offensive language classification models for zero-shot offensive span identification. To this end, we find that LIME and IG show baseline F1 of 26.35% and 44.83%, respectively. Besides, we study the effect of data set size and training process on the overall accuracy of span identification. As a result, we find both LIME and IG to show significant improvement with Masked Data Augmentation and Multilabel Training, with F1 of 50.23% and 47.38% respectively. Disclaimer : This paper contains examples that may be considered profane, vulgar, or offensive. The examples do not represent the views of the authors or their employers/graduate schools towards any person(s), group(s), practice(s), or entity/entities. Instead they are used to emphasize only the linguistic research challenges.
This paper presents the findings of the shared task on Multimodal Sentiment Analysis and Troll meme classification in Dravidian languages held at ACL 2022. Multimodal sentiment analysis deals with the identification of sentiment from video. In addition to video data, the task requires the analysis of corresponding text and audio features for the classification of movie reviews into five classes. We created a dataset for this task in Malayalam and Tamil. The Troll meme classification task aims to classify multimodal Troll memes into two categories. This task assumes the analysis of both text and image features for making better predictions. The performance of the participating teams was analysed using the F1-score. Only one team submitted their results in the Multimodal Sentiment Analysis task, whereas we received six submissions in the Troll meme classification task. The only team that participated in the Multimodal Sentiment Analysis shared task obtained an F1-score of 0.24. In the Troll meme classification task, the winning team achieved an F1-score of 0.596.
Offensive content moderation is vital in social media platforms to support healthy online discussions. However, their prevalence in code-mixed Dravidian languages is limited to classifying whole comments without identifying part of it contributing to offensiveness. Such limitation is primarily due to the lack of annotated data for offensive spans. Accordingly, in this shared task, we provide Tamil-English code-mixed social comments with offensive spans. This paper outlines the dataset so released, methods, and results of the submitted systems.
This paper presents an outline of the shared task on translation of under-resourced Dravidian languages at DravidianLangTech-2022 workshop to be held jointly with ACL 2022. A description of the datasets used, approach taken for analysis of submissions and the results have been illustrated in this paper. Five sub-tasks organized as a part of the shared task include the following translation pairs: Kannada to Tamil, Kannada to Telugu, Kannada to Sanskrit, Kannada to Malayalam and Kannada to Tulu. Training, development and test datasets were provided to all participants and results were evaluated on the gold standard datasets. A total of 16 research groups participated in the shared task and a total of 12 submission runs were made for evaluation. Bilingual Evaluation Understudy (BLEU) score was used for evaluation of the translations.
This paper presents the overview of the shared task on emotional analysis in Tamil. The result of the shared task is presented at the workshop. This paper presents the dataset used in the shared task, task description, and the methodology used by the participants and the evaluation results of the submission. This task is organized as two Tasks. Task A is carried with 11 emotions annotated data for social media comments in Tamil and Task B is organized with 31 fine-grained emotion annotated data for social media comments in Tamil. For conducting experiments, training and development datasets were provided to the participants and results are evaluated for the unseen data. Totally we have received around 24 submissions from 13 teams. For evaluating the models, Precision, Recall, micro average metrics are used.
We present our findings from the first shared task on Multi-task Learning in Dravidian Languages at the second Workshop on Speech and Language Technologies for Dravidian Languages. In this task, a sentence in any of three Dravidian Languages is required to be classified into two closely related tasks namely Sentiment Analyis (SA) and Offensive Language Identification (OLI). The task spans over three Dravidian Languages, namely, Kannada, Malayalam, and Tamil. It is one of the first shared tasks that focuses on Multi-task Learning for closely related tasks, especially for a very low-resourced language family such as the Dravidian language family. In total, 55 people signed up to participate in the task, and due to the intricate nature of the task, especially in its first iteration, 3 submissions have been received.
The social media is one of the significantdigital platforms that create a huge im-pact in peoples of all levels. The commentsposted on social media is powerful enoughto even change the political and businessscenarios in very few hours. They alsotend to attack a particular individual ora group of individuals. This shared taskaims at detecting the abusive comments in-volving, Homophobia, Misandry, Counter-speech, Misogyny, Xenophobia, Transpho-bic. The hope speech is also identified. Adataset collected from social media taggedwith the above said categories in Tamiland Tamil-English code-mixed languagesare given to the participants. The par-ticipants used different machine learningand deep learning algorithms. This paperpresents the overview of this task compris-ing the dataset details and results of theparticipants.
This paper addresses the issue of Hope Speech detection using machine learning techniques. Designing a robust model that helps in predicting the target class with higher accuracy is a challenging task in machine learning, especially when the distribution of the class labels is highly imbalanced. This study uses and compares the experimental outcomes of the different oversampling techniques. Many models are implemented to classify the comments into Hope and Non-Hope speech, and it found that machine learning algorithms perform better than deep learning models. The English language dataset used in this research was developed by collecting YouTube comments and is part of the task “ACL-2022:Hope Speech Detection for Equality, Diversity, and Inclusion”. The proposed model achieved a weighted F1-score of 0.55 on the test dataset and secured the first rank among the participated teams.
In recent years, various methods have been developed to control the spread of negativity by removing profane, aggressive, and offensive comments from social media platforms. There is, however, a scarcity of research focusing on embracing positivity and reinforcing supportive and reassuring content in online forums. As a result, we concentrate our research on developing systems to detect hope speech in code-mixed Kannada. As a result, we present DC-LM, a dual-channel language model that sees hope speech by using the English translations of the code-mixed dataset for additional training. The approach is jointly modelled on both English and code-mixed Kannada to enable effective cross-lingual transfer between the languages. With a weighted F1-score of 0.756, the method outperforms other models. We aim to initiate research in Kannada while encouraging researchers to take a pragmatic approach to inspire positive and supportive online content.
Social media is considered as a platform whereusers express themselves. The rise of social me-dia as one of humanity’s most important publiccommunication platforms presents a potentialprospect for early identification and manage-ment of mental illness. Depression is one suchillness that can lead to a variety of emotionaland physical problems. It is necessary to mea-sure the level of depression from the socialmedia text to treat them and to avoid the nega-tive consequences. Detecting levels of depres-sion is a challenging task since it involves themindset of the people which can change period-ically. The aim of the DepSign-LT-EDI@ACL-2022 shared task is to classify the social me-dia text into three levels of depression namely“Not Depressed”, “Moderately Depressed”, and“Severely Depressed”. This overview presentsa description on the task, the data set, method-ologies used and an analysis on the results ofthe submissions. The models that were submit-ted as a part of the shared task had used a va-riety of technologies from traditional machinelearning algorithms to deep learning models.It could be observed from the result that thetransformer based models have outperformedthe other models. Among the 31 teams whohad submitted their results for the shared task,the best macro F1-score of 0.583 was obtainedusing transformer based model.
This paper illustrates the overview of the sharedtask on automatic speech recognition in the Tamillanguage. In the shared task, spontaneousTamil speech data gathered from elderly andtransgender people was given for recognitionand evaluation. These utterances were collected from people when they communicatedin the public locations such as hospitals, markets, vegetable shop, etc. The speech corpusincludes utterances of male, female, and transgender and was split into training and testingdata. The given task was evaluated using WER(Word Error Rate). The participants used thetransformer-based model for automatic speechrecognition. Different results using differentpre-trained transformer models are discussedin this overview paper.
Homophobia and Transphobia Detection is the task of identifying homophobia, transphobia, and non-anti-LGBT+ content from the given corpus. Homophobia and transphobia are both toxic languages directed at LGBTQ+ individuals that are described as hate speech. This paper summarizes our findings on the “Homophobia and Transphobia Detection in social media comments” shared task held at LT-EDI 2022 - ACL 2022 1. This shared taskfocused on three sub-tasks for Tamil, English, and Tamil-English (code-mixed) languages. It received 10 systems for Tamil, 13 systems for English, and 11 systems for Tamil-English. The best systems for Tamil, English, and Tamil-English scored 0.570, 0.870, and 0.610, respectively, on average macro F1-score.
Hope Speech detection is the task of classifying a sentence as hope speech or non-hope speech given a corpus of sentences. Hope speech is any message or content that is positive, encouraging, reassuring, inclusive and supportive that inspires and engenders optimism in the minds of people. In contrast to identifying and censoring negative speech patterns, hope speech detection is focussed on recognising and promoting positive speech patterns online. In this paper, we report an overview of the findings and results from the shared task on hope speech detection for Tamil, Malayalam, Kannada, English and Spanish languages conducted in the second workshop on Language Technology for Equality, Diversity and Inclusion (LT-EDI-2022) organised as a part of ACL 2022. The participants were provided with annotated training & development datasets and unlabelled test datasets in all the five languages. The goal of the shared task is to classify the given sentences into one of the two hope speech classes. The performances of the systems submitted by the participants were evaluated in terms of micro-F1 score and weighted-F1 score. The datasets for this challenge are openly available
Social media platforms such as Twitter and Facebook have been utilised for various research studies, from the cohort-level discussion to community-driven approaches to address the challenges in utilizing social media data for health, clinical and biomedical information. Detection of medical jargon’s, named entity recognition, multi-word expression becomes the primary, fundamental steps in solving those challenges. In this paper, we enumerate the ULD-NUIG team’s system, designed as part of Social Media Mining for Health Applications (#SMM4H) Shared Task 2021. The team conducted a series of experiments to explore the challenges of task 6 and task 5. The submitted systems achieve F-1 0.84 and 0.53 score for task 6 and 5 respectively.
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.
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.
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.
Tamil is a Dravidian language that is commonly used and spoken in the southern part of Asia. During the 21st century and in the era of social media, memes have been a fun moment during the day to day life of people. Here, we try to analyze the true meaning of Tamil memes by classifying them as troll or non-troll. We present an ingenious model consisting of transformer-transformer architecture that tries to attain state of the art by using attention as its main component. The dataset consists of troll and non-troll images with their captions as texts. The task is a binary classification task. The objective of the model was to pay more and more attention to the extracted features and to ignore the noise in both images and text.
This paper demonstrates our work for the shared task on Offensive Language Identification in Dravidian Languages-EACL 2021. Offensive language detection in the various social media platforms was identified previously. But with the increase in diversity of users, there is a need to identify the offensive language in multilingual posts that are largely code-mixed or written in a non-native script. We approach this challenge with various transfer learning-based models to classify a given post or comment in Dravidian languages (Malayalam, Tamil, and Kannada) into 6 categories. The source codes for our systems are published.
This paper describes the IIITK team’s submissions to the offensive language identification, and troll memes classification shared tasks for Dravidian languages at DravidianLangTech 2021 workshop@EACL 2021. Our best configuration for Tamil troll meme classification achieved 0.55 weighted average F1 score, and for offensive language identification, our system achieved weighted F1 scores of 0.75 for Tamil, 0.95 for Malayalam, and 0.71 for Kannada. Our rank on Tamil troll meme classification is 2, and offensive language identification in Tamil, Malayalam and Kannada are 3, 3 and 4 respectively.
This paper reports the Machine Translation (MT) systems submitted by the IIITT team for the English→Marathi and English⇔Irish language pairs LoResMT 2021 shared task. The task focuses on getting exceptional translations for rather low-resourced languages like Irish and Marathi. We fine-tune IndicTrans, a pretrained multilingual NMT model for English→Marathi, using external parallel corpus as input for additional training. We have used a pretrained Helsinki-NLP Opus MT English⇔Irish model for the latter language pair. Our approaches yield relatively promising results on the BLEU metrics. Under the team name IIITT, our systems ranked 1, 1, and 2 in English→Marathi, Irish→English, and English→Irish respectively. The codes for our systems are published1 .
In a world abounding in constant protests resulting from events like a global pandemic, climate change, religious or political conflicts, there has always been a need to detect events/protests before getting amplified by news media or social media. This paper demonstrates our work on the sentence classification subtask of multilingual protest detection in CASE@ACL-IJCNLP 2021. We approached this task by employing various multilingual pre-trained transformer models to classify if any sentence contains information about an event that has transpired or not. We performed soft voting over the models, achieving the best results among the models, accomplishing a macro F1-Score of 0.8291, 0.7578, and 0.7951 in English, Spanish, and Portuguese, respectively.
Data in general encodes human biases by default; being aware of this is a good start, and the research around how to handle it is ongoing. The term ‘bias’ is extensively used in various contexts in NLP systems. In our research the focus is specific to biases such as gender, racism, religion, demographic and other intersectional views on biases that prevail in text processing systems responsible for systematically discriminating specific population, which is not ethical in NLP. These biases exacerbate the lack of equality, diversity and inclusion of specific population while utilizing the NLP applications. The tools and technology at the intermediate level utilize biased data, and transfer or amplify this bias to the downstream applications. However, it is not enough to be colourblind, gender-neutral alone when designing a unbiased technology – instead, we should take a conscious effort by designing a unified framework to measure and benchmark the bias. In this paper, we recommend six measures and one augment measure based on the observations of the bias in data, annotations, text representations and debiasing techniques.
Hope is considered significant for the well-being, recuperation and restoration of human life by health professionals. Hope speech reflects the belief that one can discover pathways to their desired objectives and become roused to utilise those pathways. To encourage research in natural language processing towards positive reinforcement approach, we created a hope speech detection dataset. This paper reports on the shared task of hope speech detection for Tamil, English, and Malayalam languages. The shared task was conducted as a part of the EACL 2021 workshop on Language Technology for Equality, Diversity, and Inclusion (LT-EDI-2021). We summarize here the datasets for this challenge which are openly available at https://competitions.codalab.org/competitions/27653, and present an overview of the methods and the results of the competing systems. To the best of our knowledge, this is the first shared task to conduct hope speech detection.
In a world with serious challenges like climate change, religious and political conflicts, global pandemics, terrorism, and racial discrimination, an internet full of hate speech, abusive and offensive content is the last thing we desire for. In this paper, we work to identify and promote positive and supportive content on these platforms. We work with several transformer-based models to classify social media comments as hope speech or not hope speech in English, Malayalam, and Tamil languages. This paper portrays our work for the Shared Task on Hope Speech Detection for Equality, Diversity, and Inclusion at LT-EDI 2021- EACL 2021. The codes for our best submission can be viewed.
This paper describes the IIITK’s team submissions to the hope speech detection for equality, diversity and inclusion in Dravidian languages shared task organized by LT-EDI 2021 workshop@EACL 2021. Our best configurations for the shared tasks achieve weighted F1 scores of 0.60 for Tamil, 0.83 for Malayalam, and 0.93 for English. We have secured ranks of 4, 3, 2 in Tamil, Malayalam and English respectively.
This paper describes the results of the shared tasks organized as part of the VarDial Evaluation Campaign 2021. The campaign was part of the eighth workshop on Natural Language Processing (NLP) for Similar Languages, Varieties and Dialects (VarDial), co-located with EACL 2021. Four separate shared tasks were included this year: Dravidian Language Identification (DLI), Romanian Dialect Identification (RDI), Social Media Variety Geolocation (SMG), and Uralic Language Identification (ULI). DLI was organized for the first time and the other three continued a series of tasks from previous evaluation campaigns.
Due to the exponential increasing reach of social media, it is essential to focus on its negative aspects as it can potentially divide society and incite people into violence. In this paper, we present our system description of work on the shared task ComMA@ICON, where we have to classify how aggressive the sentence is and if the sentence is gender-biased or communal biased. These three could be the primary reasons to cause significant problems in society. Our approach utilizes different pretrained models with Attention and mean pooling methods. We were able to get Rank 1 with 0.253 Instance F1 score on Bengali, Rank 2 with 0.323 Instance F1 score on multilingual set, Rank 4 with 0.129 Instance F1 score on meitei and Rank 5 with 0.336 Instance F1 score on Hindi. The source code and the pretrained models of this work can be found here.
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.
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.
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.
NUIG-Panlingua-KMI submission to WMT 2020 seeks to push the state-of-the-art in Similar Language Translation Task for Hindi↔Marathi language pair. As part of these efforts, we conducteda series of experiments to address the challenges for translation between similar languages. Among the 4 MT systems prepared under this task, 1 PBSMT systems were prepared for Hindi↔Marathi each and 1 NMT systems were developed for Hindi↔Marathi using Byte PairEn-coding (BPE) into subwords. The results show that different architectures NMT could be an effective method for developing MT systems for closely related languages. Our Hindi-Marathi NMT system was ranked 8th among the 14 teams that participated and our Marathi-Hindi NMT system was ranked 8th among the 11 teams participated for the task.
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.
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.
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.
Social media are interactive platforms that facilitate the creation or sharing of information, ideas or other forms of expression among people. This exchange is not free from offensive, trolling or malicious contents targeting users or communities. One way of trolling is by making memes, which in most cases combines an image with a concept or catchphrase. The challenge of dealing with memes is that they are region-specific and their meaning is often obscured in humour or sarcasm. To facilitate the computational modelling of trolling in the memes for Indian languages, we created a meme dataset for Tamil (TamilMemes). We annotated and released the dataset containing suspected trolls and not-troll memes. In this paper, we use the a image classification to address the difficulties involved in the classification of troll memes with the existing methods. We found that the identification of a troll meme with such an image classifier is not feasible which has been corroborated with precision, recall and F1-score.
A meme is a form of media that spreads an idea or emotion across the internet. As posting meme has become a new form of communication of the web, due to the multimodal nature of memes, postings of hateful memes or related events like trolling, cyberbullying are increasing day by day. Hate speech, offensive content and aggression content detection have been extensively explored in a single modality such as text or image. However, combining two modalities to detect offensive content is still a developing area. Memes make it even more challenging since they express humour and sarcasm in an implicit way, because of which the meme may not be offensive if we only consider the text or the image. Therefore, it is necessary to combine both modalities to identify whether a given meme is offensive or not. Since there was no publicly available dataset for multimodal offensive meme content detection, we leveraged the memes related to the 2016 U.S. presidential election and created the MultiOFF multimodal meme dataset for offensive content detection dataset. We subsequently developed a classifier for this task using the MultiOFF dataset. We use an early fusion technique to combine the image and text modality and compare it with a text- and an image-only baseline to investigate its effectiveness. Our results show improvements in terms of Precision, Recall, and F-Score. The code and dataset for this paper is published in https://github.com/bharathichezhiyan/Multimodal-Meme-Classification-Identifying-Offensive-Content-in-Image-and-Text
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.
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, which can simultaneously learn sentence embeddings and cluster assignments from short texts. The UDLDI model understands 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-art unsupervised methods and our model has outperformed state-of-the-art LI and DI systems in supervised settings.
Wordnets are extensively used in natural language processing, but the current approaches for manually building a wordnet from scratch involves large research groups for a long period of time, which are typically not available for under-resourced languages. Even if wordnet-like resources are available for under-resourced languages, they are often not easily accessible, which can alter the results of applications using these resources. Our proposed method presents an expand approach for improving and generating wordnets with the help of machine translation. We apply our methods to improve and extend wordnets for the Dravidian languages, i.e., Tamil, Telugu, Kannada, which are severly under-resourced languages. We report evaluation results of the generated wordnet senses in term of precision for these languages. In addition to that, we carried out a manual evaluation of the translations for the Tamil language, where we demonstrate that our approach can aid in improving wordnet resources for under-resourced Dravidian languages.