Hate Speech Detection Using Deep Learning (IJFMR)

Hate speech is a growing problem in digital spaces. The current project introduces a machine learning system that detects such content in text using NLP techniques. It leverages a labeled dataset of hate and non-hate speech and explores models like Logistic Regression, Naive Bayes, SVM, LSTM, and BERT. Text features are derived from TF-IDF and word embeddings. The model is deployed via a web interface that provides real-time feedback, helping platforms moderate offensive content. Results show strong accuracy, supporting safer online environments through AI.

https://www.ijfmr.com/papers/2025/3/47441.pdf