Survey of Computational Methods in Cyberbullying Research

One of the most pressing problems in high schools is bullying. However, with today's online and mobile technologies, bullying is moving beyond the schoolyards via cell phones, social networks, online text, video and images, etc. As bad as fighting and bullying were before the advent of the mobile Internet, the recording and posting of hurtful content online has magnified the harmful reach of bullying, enabling 24/7 bullying. Although cyberbullying may not cause any physical damage initially, it has potentially devastating psychological effects like depression, low self-esteem, suicide ideation, and even suicide. Given the gravity of the consequences cyberbullying has on its victims and its rapid spread among middle and high school students, there is an immediate and pressing need for research to understand how cyber-bullying occurs in OSNs today, so that effective techniques can be developed to accurately detect cyberbullying. Most of the previous research on cyberbullying focused on the psychological component of cyberbullying. Only recently have researchers begun to explore computational methods for detecting cyberbullying. This survey provides insight into the problem of cyberbullying in social networks by investigating existing methods in the research literature. We will first review various computational approaches in analysis, detection and prediction of cyberbullying, harassment and aggressive behavior. Then we discuss the key technical limitations and challenges researchers are facing in this field.