Cyberbullying detection and prevention: Data mining and psychological perspective

Bullying is defined as targeting an individual or a group of individuals and exposing them to ridicule and negative actions both physical and mental deliberately. This is a common but serious and demoralizing experience that every individual encounters at least once in his or her lifetime. With the advent of technology, a form of bullying known as cyberbullying has spread very quickly targeting masses of innocent people very easily. Cyberbullying involves the use of computers, mobile phones, etc. for bullying activities. In this paper we focus on the data mining and machine learning techniques which have been proposed to detect and prevent cyberbullying and implement one such machine learning technique to identify the presence or absence of cyberbullying using the dataset from a popular social networking website. We also discuss the psychological factors related to cyberbullying and how the problem can be tackled along those factors. A few proposals for the future algorithms for the detection and prevention of cyberbullying are also put forth.