Robustness analysis of privacy-preserving model-based recommendation schemes

• Published in: Expert systems with applications 2014-06, Vol.41 (8), p.3671-3681
• Main Authors: Bilge, Alper, Gunes, Ihsan, Polat, Huseyin
• Format: Article
• Language: English
• Subjects: Algorithms; Applied sciences; Collaborative filtering; Computer science; control theory; systems; Computer systems and distributed systems. User interface; Exact sciences and technology; Expert systems; Filtering; Filtration; Handles; Joining; Mathematical models; Memory and file management (including protection and security); Memory organisation. Data processing; Privacy; Recommendation; Robustness; Shilling; Software
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2013.11.039

•We examine the robustness of model-based recommendation methods with privacy.•SVD-based scheme with privacy is the most robust method against shilling attacks.•Model-based prediction methods with privacy are more robust than memory-based ones.•Segment attack is the most effective one against model-based schemes with privacy.•Increasing filler size is more effective than increasing attack size. Privacy-preserving model-based recommendation methods are preferable over privacy-preserving memory-based schemes due to their online efficiency. Model-based prediction algorithms without privacy concerns have been investigated with respect to shilling attacks. Similarly, various privacy-preserving model-based recommendation techniques have been proposed to handle privacy issues. However, privacy-preserving model-based collaborative filtering schemes might be subjected to shilling or profile injection attacks. Therefore, their robustness against such attacks should be scrutinized. In this paper, we investigate robustness of four well-known privacy-preserving model-based recommendation methods against six shilling attacks. We first apply masked data-based profile injection attacks to privacy-preserving k-means-, discrete wavelet transform-, singular value decomposition-, and item-based prediction algorithms. We then perform comprehensive experiments using real data to evaluate their robustness against profile injection attacks. Next, we compare non-private model-based methods with their privacy-preserving correspondences in terms of robustness. Moreover, well-known privacy-preserving memory- and model-based prediction methods are compared with respect to robustness against shilling attacks. Our empirical analysis show that couple of model-based schemes with privacy are very robust.