Authors:

Chloe Kilroy

Date Created:

Not specified

Course Title:

Professor:

Not specified

About Paper:

Cyberbullying continues to grow in prevalence and its impact is felt by thousands worldwide. This study seeks a Network Science perspective on cyberbullying interaction patterns on the popular social media platform, Instagram. Using a human-annotated cyberbullying dataset of 414 Instagram posts and 35,364 comments, we outline a set of heuristics for building Session Graphs, where nodes represent users and their cyberbullying roles, and edges represent their exchanged communications via comments. Users within a comment thread embody four distinct roles: Bully, Victim, Defender, and Bystander. We ignore the Bystander roles as they don't directly engage in the discourse. The three active roles are differentiated as nodes with different shapes and colors in a Session Graph. Edge weights reflect the average severity of a comment, 1 for non-cyberbullying or mild, 2 for moderate, and 3 for severe. Over these graphs, we compute the Bully Score, a measure of the net malice introduced by Bullies as they attack Victims (attacks minus pushback), and the Victim Score, a measure of the net support Victims receive from their Defenders (support minus attacks). In conjunction, these scores quantify the balance of power within each post. We find that a majority of posts have negative Victim Scores (attacks outweighing support), while the Bully Score distribution has a slight positive skew (attacks outweighing pushback). We believe this is the first study to explore granular cyberbullying network interactions using human annotations on Instagram at the post and comment levels. Our findings can help develop proactive moderation techniques to better tackle cyberbullying on social media.

Source:

Loyola University Chicago

Topics:

No topics listed

Co-authors:

Chloe Kilroy