From risk factors to detection and intervention: a practical proposal for future work on cyberbullying

Abstract

While there is an increasing flow of media stories reporting cases of cyberbullying, particularly within online social media, research efforts in the academic community are scattered over different topics across the social science and computer science academic disciplines. In this work, we explored research pertaining to cyberbullying, conducted across disciplines. We mainly sought to understand scholarly activity on intelligence techniques for the detection of cyberbullying when it occurs. Our findings suggest that the vast majority of academic contributions on cyberbullying focus on understanding the phenomenon, risk factors, and threats, with the prospect of suggesting possible protection strategies. There is less work on intelligence techniques for the detection of cyberbullying when it occurs, while currently deployed algorithms seem to detect the problem only up to some degree of success. The article summarises the current trends aiming to encourage discussion and research with a new scope; we call for more research tackling the problem by leveraging statistical models and computational mechanisms geared to detect, intervene, and prevent cyberbullying. Coupling intelligence techniques with specific web technology problems can help combat this social menace. We argue that a multidisciplinary approach is needed, with expertise from human–computer interaction, psychology, computer science, and sociology, for current challenges to be addressed and significant progress to be made.