Threat Analysis and malicious user detection in reputation systems using Mean Bisector Analysis and Cosine Similarity (MBACS)

Feedback reputation systems are gaining popularity as dealing with unfair ratings in reputation systems has been recognized as an important but difficult task. This problem is challenging when the number of true user ratings is relatively small and unfair ratings plays majority in rated values. In this paper, we propose a new method to find malicious users in online reputation systems using Mean Bisector Analysis and Cosine Similarity (MBACS). Here the effort is mainly concentrated on abnormals in both rating-values domain and the malicious users domain. MBACS is very efficient to detect malicious user ratings and aggregate trustful ratings. The proposed reputation system is evaluated through simulations, MBACS system can significantly reduce the impact of unfair ratings.