Cyberbullying Detection on tweets from Twitter using Machine Learning Algorithms

Abstract

Cyberbullying has become a serious concern in today’s society, with the widespread use of social media platforms and other online communication channels. The negative impact of cyberbullying can be severe, ranging from emotional distress to even suicide. In this paper, we explore the use of machine learning algorithms to analyze the content of messages and identify those that contain abusive or harassing language. We used six popular algorithms, namely Logistic Regression (LR), k-Nearest Neighbors (KNN), Random Forest (RF), Classification and Regression Trees (CART), Naive Bayes and Support Vector Machine (SVM), to train and test our models on a dataset of Twitter tweets. To evaluate the performance of these models, we calculated several metrics, including accuracy, F1-score, cross-validation score and receiver operating characteristic (ROC) curve. Results show that Random Forest outperformed the other algorithms with an accuracy of 94%. The results suggest that the proposed method can be an effective automated tool for detecting and preventing cyberbullying on social media platforms like Twitter.