A Survey of Shilling Attacks in Collaborative Filtering Recommender Systems

Abstract

Recommender systems have emerged in the past several years as an effective way to help people cope with the problem of information overload. However, collaborative recommender systems are known to be highly vulnerable to attacks. Attackers can inject biased profile data to have a significant impact on the recommendations produced. This paper provides a comprehensive review of shilling attack in recommender systems. We present a survey of existing research on the shilling model, algorithm dependence, attack detection, and attack evaluation metrics.