Effective Approaches for Automated Textual Based Cyberbullying Detection: A Survey

Online social networking sites are being rapidly increased in recent years. Through these social sites people all over the world can connect and they express their own feelings, emotions, interests etc. As the growth of online social networks, security is to be an important concern. Cyber crime is one of the major areas which committed in internet. Social Medias such as Facebook, Twitter are gaining more popularity as spreading messages to others. Messaging that provides opportunities to create harmful activities, named as cyberbullying. As the amount of crimes growing every day, it is impossible to perform manual detection on the dataset and extract useful information. Therefore, machine learning techniques are used, which has the ability to monitor and automatically detecting harmful online activities such as bullying messages, and helps to construct a healthy and safe social media environment. In this paper, we present a systematic review of published researches based on the automatic text-based content cyberbullying detection approaches. This paper essentially serves as a resource for researchers to determine where to best direct their future research efforts in this field.