Understanding and Fighting Bullying With Machine Learning

Bullying, in both physical and cyber worlds, has been recognized as a serious health issue among adolescents. Given its significance, scholars are charged with identifying factors that influence bullying involvement in a timely fashion. However, previous social studies of bullying are handicapped by data scarcity. The standard psychological science approach to studying bullying is to conduct personal surveys in schools. The sample size is typically in the hundreds, and these surveys are often collected only once. On the other hand, the few computational studies narrowly restrict themselves to cyberbullying, which accounts for only a small fraction of all bullying episodes.