Title: User privacy protection algorithm of perceptual recommendation system based on group recommendation
Authors: Xuefeng Ding; Xuehong Liu
Addresses: Department of Information Management Center, Sichuan University, Chengdu 610041, China ' Department of Information Management Center, Sichuan University, Chengdu 610041, China
Abstract: In order to overcome the problems of low recommendation quality and poor privacy protection ability of existing user privacy protection algorithms. This paper proposes a new privacy protection algorithm based on group recommendation. The general model of recommender system is analysed by user modelling, recommender object modelling and recommender algorithm, and the privacy concerns of perceptual recommender system are analysed by questionnaire. Based on the classification of user privacy concerns, group recommendation is introduced, and the user data is submitted anonymously through the anonymous nature of senders provided by crowds network, so as to prevent the server and malicious users from identifying good users and complete the research of protection algorithm. The experimental results show that the proposed algorithm has good recommendation quality, the highest recommendation accuracy of 99%, and high anti attack success rate and reliability.
Keywords: perceptual recommendation system; user privacy; protection; group recommendation; recommender algorithm.
International Journal of Autonomous and Adaptive Communications Systems, 2020 Vol.13 No.2, pp.135 - 150
Received: 29 Sep 2019
Accepted: 20 Nov 2019
Published online: 24 Sep 2020 *