In recently years, investigation analytics has become more important with the increasing rates of cybercrimes in the world of social media, particularly during the Covid-19 pandemic. Cyberbullying is increasing nowadays with millions of people joining social media applications. This paper is aimed to conduct a textual analysis study of cyberbullying crime in Arabic language based on selected common cases in Twitter, Facebook and Instagram applications. The analysis obtained data using one of the machine learning algorithms, which is Decision Tree (DT) algorithm. System model has been trained to identify cyberbullying crime in Arabic language. The results obtained achieved an accuracy rate of 99.08%.