In recent years, bullying and aggression against social media users have grown signi cantly, causing serious consequences to victims of all demographics. Nowadays, cyberbullying a ects more than half of young social media users worldwide, su ering from prolonged and/or coordinated digital harassment. Also, tools and technologies geared to understand and mitigate it are scarce and mostly ine ective. In this paper, we present a principled and scalable approach to detect bullying and aggressive behavior on Twitter. We propose a robust methodology for extracting text, user, and network-based attributes, studying the properties of bullies and aggressors, and what features distinguish them from regular users. We nd that bullies post less, participate in fewer online communities, and are less popular than normal users. Aggressors are relatively popular and tend to include more negativity in their posts. We evaluate our methodology using a corpus of 1.6M tweets posted over 3 months, and show that machine learning classification algorithms can accurately detect users exhibiting bullying and aggressive behavior, with over 90% AUC.
@inproceedings{Chatzakou2017Mean_Birds,
title = {{Mean Birds: Detecting Aggression and Bullying on Twitter}},
author = {Chatzakou, Despoina and Kourtellis, Nicolas and Blackburn, Jeremy and De Cristofaro, Emiliano and Stringhini, Gianluca and Vakali, Athena},
booktitle = {Proceedings of the 2017 International ACM Web Science Conference (WebSci)},
month = {June},
year = {2017},
address = {Troy, NY},
publisher = {ACM}
}