Understanding and Modeling Rumor Propagation for Vulnerability Assessment of Social Media Platforms

NSF CNS-1742847
Principle Investigators


Award Information

This website is based upon work supported by the National Science Foundation under Grant No. CNS-1742847. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

Project Information

As social media becomes a primary news source, rumors can spread widely in a short time. In recent cases, the rapid spread of misinformation caused social panic, had a dramatic financial impact, and put individuals and communities at great risk. Even when correct information is eventually disseminated, large delays can have devastating consequences. To determine how vulnerable networks are to misinformation spread, and to develop effective proactive and reactive counter-measures, it is necessary to study rumor propagation. However, rumor propagation is challenging to model and capture due to its dynamic complexity and self-sustaining nature.

The rapid diffusion of rumors across online social networks is influenced by numerous factors from both local (user forwarding behavior) and global (network diffusion) perspectives. Diffusion depends on each user's local decision about whether to propagate the information or not. That decision is related to factors including trust relationships, information provenance, and content. Diffusion also depends on the global topology of networks, how users are interconnected, as well as the rate at which users propagate the content. From this global point of view, characterizing rumor propagation across networks requires accurate yet tractable mathematical models of diffusion. This project investigates rumor diffusion via social media from these two perspectives. The integration of social psychological and computer science methodologies in this project reveals propagation patterns in large-scale networks and the psychological motivations driving user behavior. This project contributes to better monitoring, detection, and ultimately prevention of the propagation of misinformation that undermines social stability and national security. Research and training opportunities are offered to students across multiple fields, including computer science, engineering, and social science.



Yaliang Li, Chenglin Miao, Lu Su, Jing Gao, Qi Li, Bolin Ding, Zhan Qin, Kui Ren. An Efficient Two-Layer Mechanism for Privacy-Preserving Truth Discovery. ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, London, UK, August 2018, 1705-1714.


Yaqing Wang, Fenglong Ma, Zhiwei Jin, Ye Yuan, Guangxu Xun, Kishlay Jha, Lu Su, Jing Gao. EANN: Event Adversarial Neural Networks for Multi-Modal Fake News Detection. ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, London, UK, August 2018, 849-857.


Chenglin Miao, Qi Li, Lu Su, Mengdi Huai, Wenjun Jiang, Jing Gao. Attack under Disguise: An Intelligent Data Poisoning Attack Mechanism in Crowdsourcing. The Web Conference, Lyon, France, April 2018, 13-22.


Chenglin Miao, Wenjun Jiang, Lu Su, Yaliang Li, Suxin Guo, Zhan Qin, Houping Xiao, Jing Gao, Kui Ren. Privacy-Preserving Truth Discovery in Crowd Sensing Systems. ACM Transactions on Sensor Networks, accepted, September 2018.

Code & Dataset
  • EANN: "Event-adversarial Neural Networks for Multi-Modal Fake News Detection" in [KDD18]

Last updated: November 2018.