Social media often acts as breeding grounds for different forms of offensive content. For low resource languages like Tamil, the situation is more complex due to the poor performance of multilingual or language-specific models and lack of proper benchmark datasets. Based on this shared task, Offensive Language Identification in Dravidian Languages at EACL 2021, we present an exhaustive exploration of different transformer models, We also provide a genetic algorithm technique for ensembling different models. Our ensembled models trained separately for each language secured the first position in Tamil, the second position in Kannada, and the first position in Malayalam sub-tasks. The models and codes are provided.
@article{arxiv.2102.10084,
title = {Hate-Alert@DravidianLangTech-EACL2021: Ensembling strategies for Transformer-based Offensive language Detection},
author = {Debjoy Saha and Naman Paharia and Debajit Chakraborty and Punyajoy Saha and Animesh Mukherjee},
journal= {arXiv preprint arXiv:2102.10084},
year = {2021}
}
Comments
6 pages, 1 figure, 3 tables, code available at https://github.com/Debjoy10/Hate-Alert-DravidianLangTech