The Latent Community Model for Detecting Sybils in Social Networks
Download: Paper (PDF)
Date: 7 Feb 2012
Document Type: Briefing Papers
Additional Documents: Slides
Associated Event: NDSS Symposium 2012
We propose a new statistical model and associated learning algorithm for detecting Sybil attacks in online social networks, which groups the nodes in the network into closely linked communities positioned in a latent Euclidean space. Our model outperforms state of the art algorithms in simulated attacks on real network topologies.