L3Cube-MahaHate: A Tweet-based Marathi Hate Speech Detection Dataset and BERT models

TL;DR

This paper introduces L3Cube-MahaHate, the first Marathi hate speech dataset from Twitter, and evaluates deep learning models including BERT variants, demonstrating that monolingual models outperform multilingual ones in hate speech detection.

Contribution

It presents the first large-scale Marathi hate speech dataset and benchmarks various deep learning models, including monolingual BERT, for hate speech classification.

Findings

MahaBERT achieves the best classification results.

Monolingual BERT models outperform multilingual variants.

The dataset contains over 25,000 annotated tweets.

Abstract

Social media platforms are used by a large number of people prominently to express their thoughts and opinions. However, these platforms have contributed to a substantial amount of hateful and abusive content as well. Therefore, it is important to curb the spread of hate speech on these platforms. In India, Marathi is one of the most popular languages used by a wide audience. In this work, we present L3Cube-MahaHate, the first major Hate Speech Dataset in Marathi. The dataset is curated from Twitter, annotated manually. Our dataset consists of over 25000 distinct tweets labeled into four major classes i.e hate, offensive, profane, and not. We present the approaches used for collecting and annotating the data and the challenges faced during the process. Finally, we present baseline classification results using deep learning models based on CNN, LSTM, and Transformers. We explore mono-lingual and multi-lingual variants of BERT like MahaBERT, IndicBERT, mBERT, and xlm-RoBERTa and show that mono-lingual models perform better than their multi-lingual counterparts. The MahaBERT model provides the best results on L3Cube-MahaHate Corpus. The data and models are available at https://github.com/l3cube-pune/MarathiNLP .