Combating Hate Speech on Q&A Forums with Machine Learning

Online forums are becoming increasingly popular, increasing the demand for efficient ways to handle user-generated information. In order to identify hate speech in user-generated content, this project offers a question-and-answer forum that employs machine learning (ML) algorithms and Natural Language Processing (NLP) techniques. The forum is built with the MERN stack, which includes MongoDB, Express, React, and Node.js. Several NLP strategies were implemented in the project, including feature vectorization using count vectorizer, TF-IDF vectorizer, and word2vec. Logistic regression (LR), support vector machines (SVM), decision trees (DT), and naive Bayes (NB) are the ML methods used to detect hate speech. The classification report is used as an evaluation metric to assess the performance of these methods. For model training, the “Hate Speech and Offensive Language Dataset” hosted on Kaggle and freely accessible to the public is used. The findings of the experiments show that the LR model with the count vectorizer performs better than the other model and provides a maximum accuracy of 90%.