BullyNet: Detecting Cyberbullies in Online Social Networks

A machine learning-based application termed BullyNet is developed in this study to identify instances of cyberbullying on social media. The model analyzes text to identify hate speech, abusive language, and damaging content in real time by combining Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs). Support Vector Machines (SVM) and Random Forest classifiers improve prediction accuracy while an NLP pipeline based on sentiment analysis and word embeddings identifies aggressive patterns. The application enables organizations the capability to establish safer online communities by offering a valuable resource for the early detection and mitigation of cyberbullying.