DeepHateExplainer: Explainable Hate Speech Detection in Under-resourced Bengali Language

The exponential growths of social media and micro-blogging sites not only provide platforms for empowering freedom of expression and individual voices, but also enables people to express anti-social behavior like online harassment, cyberbullying, and hate speech. Numerous works have been proposed to utilize the textual data for social and anti-social behavior analysis, by predicting the contexts mostly for highly-resourced languages like English. However, some languages are under-resourced, e.g., South Asian languages like Bengali, that lack computational resources for accurate natural language processing (NLP). In this paper, we propose an explainable approach for hate speech detection from the under-resourced Bengali language, which we called DeepHateExplainer. In our approach, Bengali texts are first comprehensively preprocessed, before classifying them into political, personal, geopolitical, and religious hates, by employing the neural ensemble method of different transformer-based neural architectures (i.e., monolingual Bangla BERT-base, multilingual BERT-cased/uncased, and XLM-RoBERTa). Subsequently, important (most and least) terms are identified with sensitivity analysis and layer-wise relevance propagation (LRP), before providing human-interpretable explanations. Finally, to measure the quality of the explanation (i.e., faithfulness), we compute the comprehensiveness and sufficiency. Evaluations against machine learning (linear and tree-based models) and deep neural networks (i.e., CNN, Bi-LSTM, and Conv-LSTM with word embeddings) baselines yield F1 scores of 84%, 90%, 88%, and 88%, for political, personal, geopolitical, and religious hates, respectively, outperforming both ML and DNN baselines.