Sybil-Resilient Online Content Rating

Obtaining user feedback (using votes) is essential in ranking user-generated online content. However, any online voting system is susceptible to the Sybil attack where adversaries can out-vote real users by creating several Sybil identities. In this paper, we present SumUp, a Sybilresilient online content rating system that leverages trust networks among users to defend against Sybil attacks with strong security guarantees. SumUp addresses the basic vote aggregation problem of how to aggregate votes from different users in a trust network in the face of Sybil identities casting an arbitrarily large number of bogus votes. By using the technique of adaptive vote flow aggregation, SumUp can significantly limit the number of bogus votes cast by adversaries to no more than the number of attack edges in the trust network (with high probability). SumUp leverages user voting history to further restrict the voting power of adversaries who continuously misbehave to below the attack edges. Using detailed evaluation of several existing social networks (Digg, YouTube, Flickr, LiveJournal), we show SumUp’s ability to handle Sybil attack. By applying SumUp on the voting trace of Digg (online news voting site), we have detected strong evidence of attack on many articles marked “popular” by Digg.