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

EAGER：协作：理解和建模谣言传播以进行社交媒体平台的漏洞评估

• 批准号: 1742845
• 负责人: Jing Gao
• 金额: $ 23万
• 依托单位: SUNY at Buffalo
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-10-01 至 2020-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1742845&HistoricalAwards=false
• 关键词: EAGER; Collaborative; Understanding; Modeling; Rumor

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.

随着社交媒体成为主要新闻来源，谣言可以在短时间内广泛传播。在最近的案例中，错误信息的迅速传播引发了社会恐慌，产生了巨大的财务影响，并使个人和社区面临巨大风险。即使最终传播了正确的信息，严重的延误也可能造成灾难性的后果。为了确定网络对错误信息传播的脆弱程度，并制定有效的主动和被动对策，有必要研究谣言传播。然而，由于谣言传播的动态复杂性和自我维持的性质，对其建模和捕获具有挑战性。谣言在在线社交网络中的快速传播受到本地（用户转发行为）和全球（网络传播）角度的众多因素的影响。传播取决于每个用户关于是否传播信息的本地决定。 该决定与信任关系、信息来源和内容等因素有关。 传播还取决于网络的全局拓扑、用户如何互连以及用户传播内容的速率。 从全球角度来看，描述网络中谣言传播的特征需要准确且易于处理的传播数学模型。 本项目从这两个角度研究社交媒体上的谣言传播情况。该项目将社会心理学和计算机科学方法相结合，揭示了大规模网络中的传播模式以及驱动用户行为的心理动机。该项目有助于更好地监测、发现并最终防止破坏社会稳定和国家安全的错误信息的传播。为多个领域的学生提供研究和培训机会，包括计算机科学、工程和社会科学。

项目成果

Privacy-Preserving Truth Discovery in Crowd Sensing Systems

人群感知系统中的隐私保护真相发现

• DOI: 10.1145/3277505
• 发表时间: 2019-01-09
• 期刊: ACM Transactions on Sensor Networks (TOSN)
• 作者: Chenglin Miao;Wenjun Jiang;Lu Su;Yaliang Li;Suxin Guo;Zhan Qin;Houping Xiao;Jing Gao;Kui Ren
• 通讯作者: Kui Ren

Data Poisoning Attack against Knowledge Graph Embedding

针对知识图嵌入的数据中毒攻击

• DOI: 10.24963/ijcai.2019/674
• 发表时间: 2019-04-26
• 作者: Hengtong Zhang;T. Zheng;Jing Gao;Chenglin Miao;Lu Su;Yaliang Li;K. Ren
• 通讯作者: K. Ren

In News We Trust?: Examining Credibility and Sharing Behaviors of Fake News

在我们信任的新闻中？：检查假新闻的可信度和分享行为

• DOI: 10.1145/3328529.3328554
• 发表时间: 2019-01
• 期刊: The 10th International Conference on Social Media and Society
• 作者: Stefanone, Michael A.;Vollmer, Matthew;Covert, Jessica M.
• 通讯作者: Covert, Jessica M.

Weak Supervision for Fake News Detection via Reinforcement Learning

• DOI: 10.1609/aaai.v34i01.5389
• 发表时间: 2019-12-28
• 期刊: ArXiv
• 作者: Yaqing Wang;Weifeng Yang;Fenglong Ma;Jin Xu;Bin Zhong;Qiang Deng;Jing Gao
• 通讯作者: Jing Gao

EANN: Event Adversarial Neural Networks for Multi-Modal Fake News Detection

EANN：用于多模态假新闻检测的事件对抗神经网络

• DOI: 10.1145/3219819.3219903
• 发表时间: 2018-07-19
• 期刊: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
• 作者: Yaqing Wang;Fenglong Ma;Zhiwei Jin;Ye Yuan;Guangxu Xun;Kishlay Jha;Lu Su;Jing Gao
• 通讯作者: Jing Gao