SybilFence: Improving Social-Graph-Based Sybil Defenses with User Negative Feedback

Detecting and suspending fake accounts (Sybils) in online social networking (OSN) services protects both OSN operators and OSN users from illegal exploitation. Existing social-graph-based defense schemes effectively bound the accepted Sybils to the total number of social connections between Sybils and non-Sybil users. However, Sybils may still evade the defenses by soliciting many social connections to real users. We propose SybilFence, a system that improves over social-graph-based Sybil defenses to further thwart Sybils. SybilFence is based on the observation that even well-maintained fake accounts inevitably receive a significant number of user negative feedback, such as the rejections to their friend requests. Our key idea is to discount the social edges on users that have received negative feedback, thereby limiting the impact of Sybils' social edges. The preliminary simulation results show that our proposal is more resilient to attacks where fake accounts continuously solicit social connections over time.

[1]  Adam Tauman Kalai,et al.  A Novel Approach to Propagating Distrust , 2010, WINE.

[2]  George Danezis,et al.  SybilInfer: Detecting Sybil Nodes using Social Networks , 2009, NDSS.

[3]  Konstantin Beznosov,et al.  The socialbot network: when bots socialize for fame and money , 2011, ACSAC '11.

[4]  Feng Xiao,et al.  SybilLimit: A Near-Optimal Social Network Defense against Sybil Attacks , 2008, 2008 IEEE Symposium on Security and Privacy (sp 2008).

[5]  Christos Faloutsos,et al.  Sampling from large graphs , 2006, KDD '06.

[6]  Albert,et al.  Emergence of scaling in random networks , 1999, Science.

[7]  Michael Sirivianos,et al.  Aiding the Detection of Fake Accounts in Large Scale Social Online Services , 2012, NSDI.

[8]  Krishna P. Gummadi,et al.  An analysis of social network-based Sybil defenses , 2010, SIGCOMM '10.

[9]  Michael Kaminsky,et al.  SybilGuard: Defending Against Sybil Attacks via Social Networks , 2008, IEEE/ACM Transactions on Networking.

[10]  Michael Kaminsky,et al.  SybilGuard: defending against sybil attacks via social networks , 2006, SIGCOMM.

[11]  Alan Mislove,et al.  Bazaar: Strengthening User Reputations in Online Marketplaces , 2011, NSDI.

[12]  Jun Hu,et al.  Detecting and characterizing social spam campaigns , 2010, CCS '10.

[13]  Lakshminarayanan Subramanian,et al.  Sybil-Resilient Online Content Voting , 2009, NSDI.

[14]  Georg Lausen,et al.  Propagation Models for Trust and Distrust in Social Networks , 2005, Inf. Syst. Frontiers.

[15]  Michael Kaminsky,et al.  SybilGuard: defending against sybil attacks via social networks , 2008, TNET.

[16]  Ben Y. Zhao,et al.  Uncovering social network sybils in the wild , 2011, IMC '11.

[17]  Christian Bauckhage,et al.  The slashdot zoo: mining a social network with negative edges , 2009, WWW.

[18]  J. Hanley,et al.  The meaning and use of the area under a receiver operating characteristic (ROC) curve. , 1982, Radiology.

[19]  Paul Van Dooren,et al.  The PageTrust Algorithm: How to rank web pages when negative links are allowed? , 2008, SDM.

[20]  Ramanathan V. Guha,et al.  Propagation of trust and distrust , 2004, WWW '04.