Big Five Personality Traits and Ensemble Machine Learning to Detect Cyber-Violence in Social Media

Abstract

Cyber-violence is a largely addressed problem in e-health researches, its focus is the detection of harmful behavior from online user-generated content in order to prevent and protect victims. In this work, we show how big five personality traits are correlated to the violent behavior of the cyber-violence perpetrator. We use a set of ensemble learning algorithms with engineered features related to the vocabulary used in each Big Five personality trait namely, Agreeableness, Conscientiousness, Extraversion, Neuroticism and Openness. The findings show a significant association between the individuals’ personality state and the harmful intention. This result can be a good indicator of online users’ susceptibility to cyber-violence and therefore can help in dealing with it.