Optimize_Prime@DravidianLangTech-ACL2022: Abusive Comment Detection in Tamil

TL;DR

This paper addresses abusive comment detection in Tamil and code-mixed Tamil-English social media comments using ensemble, RNN, and Transformer models, achieving macro F1 scores around 0.43-0.45.

Contribution

It introduces an approach combining ensemble, RNN, and Transformer models for abusive comment detection in low-resource Tamil and code-mixed data, with state-of-the-art results.

Findings

MuRIL and XLM-RoBERTA achieved 0.43 F1 score on Tamil data.

MuRIL and M-BERT achieved 0.45 F1 score on code-mixed data.

Ensemble and Transformer models improve detection accuracy.

Abstract

This paper tries to address the problem of abusive comment detection in low-resource indic languages. Abusive comments are statements that are offensive to a person or a group of people. These comments are targeted toward individuals belonging to specific ethnicities, genders, caste, race, sexuality, etc. Abusive Comment Detection is a significant problem, especially with the recent rise in social media users. This paper presents the approach used by our team - Optimize_Prime, in the ACL 2022 shared task "Abusive Comment Detection in Tamil." This task detects and classifies YouTube comments in Tamil and Tamil- English Codemixed format into multiple categories. We have used three methods to optimize our results: Ensemble models, Recurrent Neural Networks, and Transformers. In the Tamil data, MuRIL and XLM-RoBERTA were our best performing models with a macro-averaged f1 score of 0.43. Furthermore, for the Code-mixed data, MuRIL and M-BERT provided sub-lime results, with a macro-averaged f1 score of 0.45.