Detecting Arabic Cyberbullying Tweets Using Machine Learning

The advancement of technology has paved the way for a new type bullying, which often leads to negative stigma in social setting. Cyberbullying is cybercrime wherein one individual becomes target harassment and hatred. It recently become more prevalent due rise usage media platforms, and, some severe situations, it even led victims’ suicides. In literature, several cyberbullying detection methods are proposed, but they mainly focused on word-based data user account attributes. Furthermore, most them related English language. Meanwhile, only few papers have studied Arabic platforms. This paper, therefore, aims use machine learning language automatic detection. proposed mechanism identifies using Support Vector Machine (SVM) classifier algorithm by real dataset obtained from YouTube Twitter train test classifier. Moreover, we include Farasa tool overcome text limitations improve bullying attacks.