Bengali Cyberbullying Detection Comments Dataset

This dataset contains 44,001 Bengali comments, curated to detect cyberbullying using Natural Language Processing (NLP) techniques. Each comment is labeled by experts, categorizing different forms of harassment and offensive behavior. The dataset enables the identification of inappropriate content, ranging from mild to severe harassment, facilitating precise classification and analysis. This resource is designed for researchers and developers working on cyberbullying detection, sentiment analysis, and content moderation in Bengali text.

If you want to work further on this dataset, cite this paper - https://arxiv.org/abs/2106.04506