Privacy-preserving collaborative recommendations based on random perturbations

详细信息    查看全文

• 作者：Nikolaos Polatidis^a ; ^{npolatidis@uom.edu.gr} ; Christos K. Georgiadis^a ; ^{geor@uom.edu.gr} ; Elias Pimenidis^b ; ^{Elias.Pimenidis@uwe.ac.uk} ; Haralambos Mouratidis^c ; ^{H.Mouratidis@brighton.ac.uk}
• 关键词：Collaborative filtering ; Random perturbations ; Multi-level privacy ; Recommender systems
• 刊名：Expert Systems with Applications
• 出版年：2017
• 出版时间：1 April 2017
• 年：2017
• 卷：71
• 期：Complete
• 页码：18-25
• 全文大小：1002 K
• 卷排序：71

文摘

We propose a solution to the privacy problem found in collaborative filtering. Our proposed method is based on multiple levels. We have evaluated our method using five real datasets and well known metrics.