Detection of Abusive Tweets Using Support Vector Machine

Twitter (X) is one of the most popular social networks in the Arab region, ranked tenth among social media applications in attracting visitors. The diversity of users and their varying interests has led to the presence of spammers posting offensive and harmful content to Twitter (X) users. This paper focuses on detecting spam messages written in Arabic. The Arabic language is based on consistent roots: past, present, and imperative. The proposed method for detecting abusive tweets employs the Support Vector Machine (SVM) to extract features. The results of Recall, Precision, F-measure, and Accuracy are used to evaluate the method's performance. In the best case, the Recall achieved is 0.995.