Design And Implementation Of Privacy-preserving Recommender System

With the rapid proliferation of recommender systems,they can help us cope with information overload wonderfully.However,privacy risks of recommender systems have caused increasing attention.Users' private data are often collected by probably untrusted recommender system in order to provide high-quality recommendation.Meanwhile,malicious attackers may utilize recommendation results to make inferences about other users' private data.The problems above,if not handled well,will definitely lead to more serious private problems.In this paper,we work on providing privacy preserving techniques and methods in recommender systems.In addition,we design and implement a privacy preserving social point-of-interest recommender system called PPS-POI-Rec,which implements our algorithms for recommender system.Specifically,our work covers the following several aspects:(1)We present a secure and efficient framework for privacy preserving social recommendation.Our framework is built on mature cryptographic building blocks,including Paillier cryptosystem and Yao's protocol,which lays a solid foundation for the security of our framework.We theoretically prove the security and analyze the complexity of our framework.Empirical study shows our framework has a linear complexity with respect to the number of users and items in recommender systems and is practical in real application.(2)We propose a hybrid approach for privacy preserving recommendation by combining differential privacy with randomized perturbation.We theoretically show the noise added by randomized perturbation has limited effect on recommendation accuracy and the noise added by differential privacy can be well controlled based on the sensitivity analysis of functions on the perturbed data.Extensive experiments on three real world datasets are implemented.The hybrid approach generally provides more privacy protection with acceptable recommendation accuracy loss.(3)We design and implement a privacy preserving social point-of-interest recommender system called PPS-POI-Rec,which implements our algorithms for recommender system.We demonstrate how to recommend the POIs to users by our privacy preserving algorithms using mobile phones,to further illustrate the significance and the practical value of our system.In this article,we study the problem of privacy protection in recommender systems and implement a prototype system that handles the problem well.The PPS-POI-Rec makes contributions in practical aspect.