Robust Sybil attack defense with information level in online Recommender Systems

摘要: As the major function of Recommender Systems (RSs) is recommending commercial items to potential consumers (i.e., system users), providing correct information RS crucial both providers and users. The influence over Online Social Networks (OSNs) expanding rapidly, whereas malicious users continuously try attack RSs with fake identities Sybils) by manipulating in adversely. In this paper, we propose a novel robust recommendation algorithm called RobuRec which exploits distinctive feature, admission control. provides highly trusted results since predicts appropriate recommendations regardless whether ratings are given honest or Sybils thanks power To demonstrate performance RobuRec, have conducted extensive experiments various datasets as well diverse scenarios. evaluation confirm that outperforms comparable schemes such PCA LTSMF significantly terms Prediction Shift (PS) Hit Ratio (HR).