Benchmarking TF-IDF, Paragraph2Vec, and DistilBERT for hate speech detection in online comments — code + research paper (ACL/arXiv style).

Framework for offensive-Language-Detection using Transformer and BiLSTM using two bench mark dataset - OLID and HateXplain.

XAI-powered transformer classifier for joint hate speech and fake news detection — fine-tunes DistilBERT on HateXplain with SHAP, LIME, and ELI5 for token-level explainability. Peer-reviewed research.

Identity Abuse / Targeted Harassment Analyzer using HateXplain dataset