Defense against sybil attacks in directed social networks

In this paper, we attempt to solve the problem of defense against sybil attacks in directed social networks. We propose a set of measures for the quality of network partitions, with modularity as a special case. We present an algorithm based on the set of measures and iterative optimization to detect the sybil region. The algorithm is evaluated using a subset of real-world social topology and is confirmed to be efficient for solving the problem. Moreover, a comparison between the proposed algorithm and SybilDefender is provided, which shows that the proposed algorithm is superior for the sybil region detection problem in directed social networks.