Prediction of Cyberbullying Incidents on the Instagram Social Network

TL;DR

This paper investigates new methods for detecting and predicting cyberbullying on Instagram by analyzing images, comments, and social features, utilizing crowdsourced labeling and classifier evaluation.

Contribution

It introduces a comprehensive approach combining media content analysis, social features, and machine learning to automatically detect cyberbullying incidents on Instagram.

Findings

Identified key features associated with cyberbullying

Developed classifiers with promising detection accuracy

Provided insights into cyberbullying patterns on Instagram

Abstract

Cyberbullying is a growing problem affecting more than half of all American teens. The main goal of this paper is to investigate fundamentally new approaches to understand and automatically detect and predict incidents of cyberbullying in Instagram, a media-based mobile social network. In this work, we have collected a sample data set consisting of Instagram images and their associated comments. We then designed a labeling study and employed human contributors at the crowd-sourced CrowdFlower website to label these media sessions for cyberbullying. A detailed analysis of the labeled data is then presented, including a study of relationships between cyberbullying and a host of features such as cyberaggression, profanity, social graph features, temporal commenting behavior, linguistic content, and image content. Using the labeled data, we further design and evaluate the performance of classifiers to automatically detect and pre- dict incidents of cyberbullying and cyberaggression.