Guilt-by-Constellation: Fraud Detection by Suspicious Clique Memberships
作者： Veronique Van Vlasselaer , Leman Akoglu , Tina Eliassi-Rad , Monique Snoeck , Bart Baesens
关键词:
摘要: Given a labeled graph containing fraudulent and legitimate nodes, which nodes group together? How can we use the riskiness of node groups to infer future label for new members group? This paper focuses on social security fraud where companies are linked resources they share. The primary goal in is detect that intentionally fail pay their contributions government. We aim by (1) propagating time-dependent exposure score each based its relationships known network, (2) deriving cliques resources, labeling these terms bankruptcy involvement, (3) characterizing company using combination intrinsic relational features membership suspicious cliques. show clique-based boost performance traditional models.
kuleuven.be
eliassi.org 参考文章(17)
Leman Akoglu, Mary McGlohon, Christos Faloutsos, OddBall: spotting anomalies in weighted graphs knowledge discovery and data mining. ,vol. 6119, pp. 410- 421 ,(2010) , 10.1007/978-3-642-13672-6_40
Rajeev Motwani, Terry Winograd, Lawrence Page, Sergey Brin, The PageRank Citation Ranking : Bringing Order to the Web the web conference. ,vol. 98, pp. 161- 172 ,(1999)
Mark Newman, Networks: An Introduction ,(2010)
Danai Koutra, Tai-You Ke, U. Kang, Duen Horng Chau, Hsing-Kuo Kenneth Pao, Christos Faloutsos, Unifying guilt-by-association approaches: theorems and fast algorithms european conference on machine learning. pp. 245- 260 ,(2011) , 10.1007/978-3-642-23783-6_16
Corinna Cortes, Daryl Pregibon, Chris Volinsky, Communities of Interest intelligent data analysis. ,vol. 6, pp. 105- 114 ,(2001) , 10.1007/3-540-44816-0_11
Ran Raz, On the complexity of matrix product symposium on the theory of computing. pp. 144- 151 ,(2002) , 10.1145/509907.509932
S BOCCALETTI, V LATORA, Y MORENO, M CHAVEZ, D HWANG, Complex networks: Structure and dynamics Physics Reports. ,vol. 424, pp. 175- 308 ,(2006) , 10.1016/J.PHYSREP.2005.10.009
Andrew Fast, Lisa Friedland, Marc Maier, Brian Taylor, David Jensen, Henry G. Goldberg, John Komoroske, Relational data pre-processing techniques for improved securities fraud detection Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '07. pp. 941- 949 ,(2007) , 10.1145/1281192.1281293
Véronique Van Vlasselaer, Jan Meskens, Dries Van Dromme, Bart Baesens, Using social network knowledge for detecting spider constructions in social security fraud advances in social networks analysis and mining. pp. 813- 820 ,(2013) , 10.1145/2492517.2500292
Brian Gallagher, Hanghang Tong, Tina Eliassi-Rad, Christos Faloutsos, Using ghost edges for classification in sparsely labeled networks Proceeding of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD 08. pp. 256- 264 ,(2008) , 10.1145/1401890.1401925