Session-Based Cyberbullying Detection: Problems and Challenges
Lu Cheng, Yasin N. Silva, Deborah L. Hall, Huan Liu
Abstract
Cyberbullying has become one of the most pressing online risks for young people, due in part to the rapid increase in social media use, and has raised serious concerns in society. Existing studies have examined various approaches to cyberbullying detection focusing on a single piece of text, whereas relatively little is known about cyberbullying detection within a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">social media session</i> . A social media session typically consists of an initial post, images/videos, a sequence of comments that involves user interactions, user information, spatial location, and other social content. By investigating cyberbullying at the level of social media sessions, researchers can draw on data that are more complex, diverse, and crucial for understanding two defining characteristics of cyberbullying, in particular: <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">repetitive acts</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">power imbalance</i> . This article thus highlights the importance of studying session-based cyberbullying detection, identifies core challenges, and serves as a resource to help direct future research efforts.