The impact of data obfuscation on the accuracy of collaborative filtering

被引：32
作者：
Berkovsky, Shlomo [1 ]
Kuflik, Tsvi [2 ]
Ricci, Francesco [3 ]
机构：
[1] CSIRO, Informat & Commun Technol Ctr, Marsfield, NSW, Australia
[2] Univ Haifa, Dept Informat Syst, IL-31999 Haifa, Israel
[3] Free Univ Bozen Bolzano, Fac Comp Sci, Bozen Bolzono, Italy
来源：
EXPERT SYSTEMS WITH APPLICATIONS | 2012年 / 39卷 / 05期
关键词：
Collaborative filtering; Recommender systems; Accuracy; Data obfuscation;
D O I：
10.1016/j.eswa.2011.11.037
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
Collaborative filtering (CF) is a widely-used technique for generating personalized recommendations. CF systems are typically based on a central storage of user profiles, i.e., the ratings given by users to items. Such centralized storage introduces potential privacy breach, since all the user profiles may be accessible by untrusted parties when breaking the access control of the centralized system. Hence, recent studies have focused on enhancing the privacy of CF users by distributing their user profiles across multiple repositories and obfuscating the user profiles to partially hide the actual user ratings. This work combines these two techniques and investigates the unavoidable side effect of data obfuscation: the reduction of the accuracy of the generated CF predictions. The evaluation, which was conducted using three different datasets, shows that considerable parts of the user profiles can be modified without observing a substantial decrease of the CF prediction accuracy. The evaluation also indicates what parts of the user profiles are required for generating accurate CF predictions. In addition, we conducted an exploratory user study that reveals positive attitude of users towards the data obfuscation. Crown Copyright (C) 2011 Published by Elsevier Ltd. All rights reserved.
引用
收藏
页码：5033 / 5042
页数：10
相关论文
共 27 条
[1]  
Ackerman M.S., 1999, Proceedings of the 1st ACM Conference on Electronic Commerce EC, P1, DOI [10.1145/336992.336995, DOI 10.1145/336992.336995]
[2]  
Agrawal R., 2004, P ACM SIGMOD INT C M, P563
[3]   A survey of peer-to-peer content distribution technologies [J].
Androutsellis-Theotokis, S ;
Spinellis, D .
ACM COMPUTING SURVEYS, 2004, 36 (04) :335-371
[4]  
[Anonymous], 2002, Peer-to-Peer Computing
[5]  
Berkvosky S, 2007, RECSYS 07: PROCEEDINGS OF THE 2007 ACM CONFERENCE ON RECOMMENDER SYSTEMS, P9
[6]  
Brier S., 1997, KEEP YOUR PRIVACY BA
[7]   Collaborative filtering with privacy [J].
Canny, J .
2002 IEEE SYMPOSIUM ON SECURITY AND PRIVACY, PROCEEDINGS, 2002, :45-57
[8]  
CRANOR L, 1999, 9943 AT T LAB RES
[9]   Eigentaste: A constant time collaborative filtering algorithm [J].
Goldberg, K ;
Roeder, T ;
Gupta, D ;
Perkins, C .
INFORMATION RETRIEVAL, 2001, 4 (02) :133-151
[10]   Evaluating collaborative filtering recommender systems [J].
Herlocker, JL ;
Konstan, JA ;
Terveen, K ;
Riedl, JT .
ACM TRANSACTIONS ON INFORMATION SYSTEMS, 2004, 22 (01) :5-53
123