SybilTrap: A graph-based semi-supervised Sybil defense scheme for online social networks

摘要

Sybil attacks are increasingly prevalent in online social networks. A malicious user canrngenerate a huge number of fake accounts to produce spam, impersonate other users, commitrnfraud, and reach many legitimate users. For security reasons, such fake accounts have to berndetected and deactivated immediately. Various defense schemes have been proposed to dealrnwith fake accounts. However, most identify fake accounts using only the structure of socialrngraphs, leading to poor performance. In this paper, we propose a new and scalable defensernscheme, SybilTrap. SybilTrap uses a semi‐supervised technique that automatically integratesrnthe underlying features of user activities with the social structure into one system. Unlikernother machine learning–based approaches, the proposed defense scheme works on unlabeledrndata, and it is effective in detecting targeted attacks, because it manipulates different levels ofrnfeatures of user profiles. We evaluate SybilTrap on a dataset collected from Twitter. We showrnthat our proposed scheme is able to accurately detect Sybil nodes as well as huge conspiraciesrnamong them.