top of page

MALICIOUS BEHAVIOR IN NETWORKED POPULATIONS

A specific recent focus of ours is disinformation in social media. Understanding that big data and AI are powering increasingly precise and widespread propaganda, we raise technical and ethical questions around detection and mitigation of subversive content.   

More broadly, we work to learn the dynamic signatures of abusive behavior in social computing systems. Areas of interest include cyberbullying, vulnerability and exploitation, and malicious actors in crowd work platforms. In this thread of research, we marry supervised machine learning methods, computer vision, natural language processing and game-theoretic models of peer influence to characterize deviant behavior in networked populations.

Through work funded by the Office of Naval Research, we are developing algorithms to categorize and respond to classes of malicious accounts on Twitter. 

​

  • P. Venkit, T. Chakravorti, V. Gupta, H. Biggs, M. Srinath, K. Gouswami, S. Rajtmajer, S. Wilson. An Audit on the Perspectives and Challenges of Hallucinations in NLP. ACL Conference on Empirical Methods in Natural Language Processing (EMNLP), November 2024. 

  • X. Wang, J. Li, S. Rajtmajer. Inside the echo chamber: Linguistic underpinnings of misinformation on Twitter. ACM Conference on Web Science (WebSci), May 2024.

  • M. Merhi, D. Lee, S. Rajtmajer. Information Operations in Turkey: Manufacturing Resilience with Free Twitter Accounts. AAAI Conference on Web and Social Media (ICWSM), June 2023.

  • X. Wang, J. Li, E. Srivatsavaya, S. Rajtmajer. Evidence of inter-state coordination amongst state-backed information operations. Scientific reports, 13(1), 7716, May 2023..

  • X. Wang, M. Wu, S. Rajtmajer. From Yellow Peril to Model Minority: Asian stereotypes in social media during the COVID-19 pandemic. ACM Conference on Web Science (WebSci), April 2023.

  • S. Rajtmajer and D. Susser. Automated Influence and the Challenge of Cognitive Security. ACM Symposium on the Science of Security (HoTSoS), February 2020.

  • C. Qiu, A. Squicciarini, S. Rajtmajer. Rating Mechanisms for Sustainability of Crowdsourcing Platforms. ACM Conference on Information and Knowledge Management (CIKM), November 2019.

  • C. Qiu, A. Squicciarini, S. Rajtmajer and J. Caverlee. Dynamic Contract Design for Heterogeneous Workers in Crowdsourcing for Quality Control. International Conference on Digital Society (IEEE ICDS), July 2017.

  • A. Squicciarini, S. Rajtmajer and C. Griffin. Positive and Negative Behavioral Analysis in Social Networks. WIRE Data Mining and Knowledge Discovery, May 2017.

  • H. Zhong, H. Li, A. Squicciarini, S. Rajtmajer, C. Griffin, D. Miller and C. Caragea. Content- Driven Detection of Cyberbullying on the Instagram Social Network. International Conference on Artificial Intelligence (IJCAI), July 2016.

  • C. Liao, A. Squicciarini, C. Griffin and S. Rajtmajer. A Hybrid Epidemic Model for Deindividuation and Antinormative Behavior in Online Social Networks. Social Network Analysis and Mining (SNAM), January 2016.

  • C. Liao, A. Squicciarini, C. Griffin and S. Rajtmajer. A Hybrid Epidemic Model for the Spread of Abusive Content in Online Social Sites. In Proc. 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), August 2015. 

  • A. Squicciarini, S. Rajtmajer, Y. Liu and C. Griffin. Identification and Characterization of Cyberbullying Dynamics in an Online Social Network. In Proc. 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), August 2015. 

  • S. Rajtmajer, C. Griffin, D. Mikesell and A. Squicciarini. An Evolutionary Game Model for the Spread of Non-Cooperative Behavior in Online Social Networks. In Proc. 30th ACM Symposium on Applied Computing, April 2015.

bottom of page