Tutorial information

Bullying, once limited to physical spaces (e.g., schools, workplaces or sports fields) and particular times of the day (e.g., school hours), can now occur anytime, anywhere. Cyberbullying can take many forms; however, it typically refers to repeated and hostile behavior performed in an effort to intentionally and repeatedly harass or harm individuals. The consequences of cyberbullying can be devastating.

The primary motivation of this tutorial is to provide a systematic overview of the state–of–the–art computational approaches for the detection, quantification, and mitigation of cyberbullying behavior in online social media. The goal is threefold:

• Discuss and clarify the notion of cyberbullying, and the related nomenclature, by drawing from the social and psychological sciences.
• Summarize the progress of the WebSci community on an emerging research area with a growing body of research and plenty of open questions, both for researchers and practitioners.
• Identify challenges associated with understanding, predicting, and preventing cyberbulling behavior.
• Discuss opportunities for future research directions

Throughout the tutorial, we highlight the open challenges that need to be addressed in understanding, predicting, and preventing cyberbullying behavior, as well as promising research directions for researchers interested in this area.

The tutorial will be held as part of the 13th ACM Web Science Conference on June 21st, 2021. A previous version of the tutorial was delivered at ICWSM in Stanford, California, USA on June 25th, 2018. Relevant materials can be found here.

Tutorial Outline and Slides

• Introduction
• Definition of cyberbullying behavior and overview of corresponding activities
• Ideal cyberbullying detection system and challenges
• Open and/or accessible datasets
• Characterization of cyberbullying behavior
• Causes of cyberbullying
• Behavioral characteristics of bullies, victims, and bystanders
• Prominent features for cyberbullying characterization based on user profiles, messages, images and videos, and network structure
• Cyberbullying detection and prediction
• Detection and prediction of cyberbullying with linguistic, visual, profile, demographic, and activity features
• Cyberbullying detection with network structure properties
• Session based, contextual and multimodal methods
• Temporal patterns of cyberbullying
• Mitigation strategies
• Education of users on the negative consequences to victims, recognition and intervention
• Promotion of positive digital behavior
• Victim support
• Bystander engagement
• Evaluation of mitigation tools
• Real–world challenges from the industry perspective
• Acquisition of high–fidelity annotated data
• Scalability
• Beyond accuracy metrics
• Conclusion

Who should attend

This tutorial targets academic, industry and government researchers and practitioners with interest in user behavior modeling, social media and social network analysis, graph mining, and detection and prediction algorithms. Beginners will learn the basics of state–of–the–art methods in this area, whereas experts in the area will have a chance for an in–depth review of both platform–specific techniques and platform–independent algorithms to detect online bullying behavior. As a result, this tutorial should appeal to researchers of several disciplines.

Biographies

Charalampos Chelmis is an Assistant Professor of Computer Science at the University at Albany, State University of New York, where he leads the Intelligent Big Data Analytics, Applications, and Systems group. He specializes in socially important data science with a proven track record on analysis, modeling, and accurate prediction of process dynamics on large–scale real–world networks. Before joining the University at Albany, he was Senior Research Associate at the University of Southern California. He obtained his Ph.D. from the University of Southern California in 2013. He is program committee member of international conferences including but not limited to AAAI, the WebConf (formerly WWW), ICWSM, and ASONAM. He is also reviewer for journals including PLOS ONE, IEEE TKDE, IEEE TCSS, and IEEE TIFS.

Daphney-Stavroula Zois is an Assistant Professor in Electrical and Computer Engineering at the University at Albany, State University of New York. She specializes in signal processing, control and machine learning for continuous and resource–efficient sensing, estimation and decision–making that adapts to unknown and uncertain environments. She has published numerous conference and journal refereed articles on the design of estimation and sequential decision–making frameworks and algorithms for a plethora of applications including e–health, government 2.0, and cyberbullying. Before joining UAlbany, she was a Postdoctoral Researcher at the University of Illinois at Urbana Champaign. She obtained her Ph.D. from the University of Southern California in 2014. Dr. Zois serves as a reviewer of high–impact journals such as IEEE TAI, IEEE TSP, IEEE TAC, ACM TSN, and conferences such as AAAI, NIPS, ICASSP, and ISIT.