A Comprehensive Study of Shilling Attacks in Recommender Systems
International Journal of Computer Science Issues · 2017
With the abundance of data available, it becomes difficult to distinguish useful information from massive amount of information available.Recommender systems serves the purpose of filtering information to provide relevant information to users that best acknowledge their needs.In order to generate efficient recommendations to its target users, a recommender system may use user data such as user identity, demographic profile, purchase history, rating history, browsing behavior etc.This may raise security and privacy concerns for a user.The goal of this paper is to address various security and privacy issues in a recommender system.In this paper, we also discuss some of the evaluation metrics for various attack models.