Method and system for predicting victim users and detecting fake user accounts in online social networks

作者: Yazan Boshmaf , Georgios Siganos , Dionysios Logothetis

DOI:

关键词:

摘要: A system and method for predicting victims detecting fake accounts in OSNs, comprising: a feature-based classifier by classifying, with classification probability, target variable of each user the OSN social graph; graph-transformer transforming graph into defense reassigning edge weights to incorporate victim predictions using graph-based detector users computing through power iteration probability random walk land on node after 0(log n) steps, assigning rank value equal node's landing normalized degree, sorting nodes their estimating detection threshold such that whose is smaller than flagged as representing account.

参考文章(8)
Sepandar D. Kamvar, Mario T. Schlosser, Method and system for reputation management in peer-to-peer networks ,(2004)
Prashun Purkayastha, Anandamoy Roychowdhary, Gaming platform utilizing a fraud detection platform ,(2013)
Pranam Kolari, Zhaohui Zheng, Anlei Dong, Yi Chang, Ruiqiang Zhang, Jing Bai, Ranking of search results based on microblog data ,(2010)
Qiang Cao, Michael Sirivianos, Xiaowei Yang, Method and system for detecting fake accounts in online social networks ,(2013)
Danny B. Lange, Hung-Chih Yang, Xiong Zhang, Identifying influential users of a social networking service ,(2012)
Travis E. Dawson, Hui Zang, Telecom fraud detection using social pattern ,(2010)