Cybercrime detection in online communications: The experimental case of cyberbullying detection in the Twitter network

The popularity of online social networks has created massive social communication among their users and this leads to a huge amount of user-generated communication data. In recent years, Cyberbullying has grown into a major problem with the growth of online communication and social media. Cyberbullying has been recognized recently as a serious national health issue among online social network users and developing an efficient detection model holds tremendous practical significance. In this paper, we have proposed set of unique features derived from Twitter; network, activity, user, and tweet content, based on these feature, we developed a supervised machine learning solution for detecting cyberbullying in the Twitter. An evaluation demonstrates that our developed detection model based on our proposed features, achieved results with an area under the receiver-operating characteristic curve of 0.943 and an f-measure of 0.936. These results indicate that the proposed model based on these features provides a feasible solution to detecting Cyberbullying in online communication environments. Finally, we compare result obtained using our proposed features with the result obtained from two baseline features. The comparison outcomes show the significance of the proposed features.