indicnlp@kgp at DravidianLangTech-EACL2021: Offensive Language Identification in Dravidian Languages
This work addresses the problem of detecting offensive content in low-resource Dravidian languages for NLP applications, but it is incremental as it applies existing methods to new datasets.
The paper tackled offensive language identification in three code-mixed Dravidian languages by using an ensemble of AWD-LSTM, BERT, and RoBERTa models, achieving weighted-average F1 scores of 0.97, 0.77, and 0.72 on Malayalam-English, Tamil-English, and Kannada-English datasets, ranking 1st, 2nd, and 3rd respectively.
The paper presents the submission of the team indicnlp@kgp to the EACL 2021 shared task "Offensive Language Identification in Dravidian Languages." The task aimed to classify different offensive content types in 3 code-mixed Dravidian language datasets. The work leverages existing state of the art approaches in text classification by incorporating additional data and transfer learning on pre-trained models. Our final submission is an ensemble of an AWD-LSTM based model along with 2 different transformer model architectures based on BERT and RoBERTa. We achieved weighted-average F1 scores of 0.97, 0.77, and 0.72 in the Malayalam-English, Tamil-English, and Kannada-English datasets ranking 1st, 2nd, and 3rd on the respective tasks.