Optimal Online Cyberbullying Detection

Cyberbullying has emerged as a serious societal and public health problem that demands accurate methods for the detection of cyber-bullying instances in an effort to mitigate the consequences. While techniques to automatically detect cyberbullying incidents have been developed, the scalability and timeliness of existing cyberbullying detection approaches have largely been ignored. We address this gap by formulating cyberbullying detection as a sequential hypothesis testing problem. Based on this formulation, we propose a novel algorithm designed to reduce the time to raise a cyberbullying alert by drastically reducing the number of feature evaluations necessary for a decision to be made. We demonstrate the effectiveness of our approach using a real-world dataset from Twitter, one of the top five networks with the highest percentage of users reporting cyberbullying instances. We show that our approach is highly scalable while not sacrificing accuracy for scalability.

1. Introduction

Bullying, once limited to physical spaces (e.g., schools, workplaces or sports fields) and particular times of the day (e.g., school hours), can now occur anytime, anywhere. Cyberbullying can take many forms, however, it typically refers to repeated and hostile behavior (e.g., hurtful comments, videos and images) performed in an effort to intentionally and repeatedly harass or harm individuals [1]. The consequences can be devastating: learning difficulties, psychological suffering and isolation, escalated physical confrontations, suicide. Statistics are staggering: over half of adolescents have engaged in or have been cyberbullied, while 10% – 20% experience it daily¹

Bullying Statistics: http://www.bullyingstatistics.org/category/bullying-statistics

.