Publication

Abstract : The research being presented here looks into the problems related to cyberbullying using data from Twitter and Wikipedia. To try and understand why people behave in this way online, the research included several challenging techniques, including a Support Vector Machine, Random Forest, Gaussian Naive Bayes, and a Voting Classifier that combines Logistic Regression, Random Forest, and Gaussian Naive Bayes. The accuracy of the Voting Classifier was 92.86% after Twitter data analysis. This demonstrates how well it detects and comprehends abusive tendencies. Not to mention, the Wikipedia dataset showed strong performance from the Voting Classifier, which had a success rate of 97.12%. Compared to other methods of research, it performed better. The research shows how machine learning algorithms can recognize and determine the identity of individuals attempting to cause harm to others via the internet