Efficient Detection of Cyberbullying in Social Media Platform

The prevalence of cyberbullying on messaging apps like WhatsApp is a major concern nowadays. The first step in resolving this problem is developing machine learning models to automatically identify the cases of cyberbullying. In order to detect cyberbullying on the messaging app like WhatsApp, this study introduces a machine learning approach that makes use of logistic regression and support vector machines (SVMs). Collecting a dataset of WhatsApp chats, including both cyberbullying and non-bullying ones, is the first step in the proposed method. In order to clean up the data and remove any noise, text preparation methods are used. Feature extraction is a powerful tool used for capturing important environmental and vocal cues related to cyberbullying. Classification models are trained by utilizing the selected features with the help of logistic regression and support vector machines. The outcomes demonstrate the efficacy of the suggested method. While Support Vector Machines (SVMs) get up to 85% accuracy, logistic regression models have 83% accuracy when it comes to cyberbullying case classification. The results show that the machine learning algorithms were successful in detecting cyberbullying in WhatsApp conversations.