Cyberbullying Detection on Social Media Using Machine Learning

Social media platforms have seen an increase in the prevalence of cyberbullying. Making social media platforms safe from cyberbullying is essential, given the popularity and extensive use of social media among people of all ages. This study compares three machine learning algorithms, Support Vector Machine (SVM), Na¨ive Bayes, and a Bidirectional Long Short-Term Memory (Bi-LSTM) on a cyberbullying Twitter dataset. Regarding the experimental results, Bi-LSTM model performs the best, achieving 98% accuracy, followed by SVM with 97% accuracy, and Naive Bayes with 85%. It shows that machine learning techniques are effective in exposing cyberbullying, and Bi-LSTM is superior to the other two traditional machine learning classifiers in our study.