File Download
  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Differential privacy for collaborative filtering recommender algorithm

TitleDifferential privacy for collaborative filtering recommender algorithm
Authors
KeywordsRecommendation
Collaborative Filtering
Inference Attack
Differential Privacy
Issue Date2016
PublisherACM.
Citation
The 6th ACM International Workshop on Security and Privacy Analytics (IWSPA 2016), New Orleans, LA., 9-11 March 2016. In Conference Proceedings, 2016, p. 9-16 How to Cite?
AbstractCollaborative filtering plays an essential role in a recommender system, which recommends a list of items to a user by learning behavior patterns from user rating matrix. However, if an attacker has some auxiliary knowledge about a user purchase history, he/she can infer more information about this user. This brings great threats to user privacy. Some methods adopt differential privacy algorithms in collaborative filtering by adding noises to a rating matrix. Although they provide theoretically private results, the influence on recommendation accuracy are not discussed. In this paper, we solve the privacy problem in recommender system in a different way by applying the differential privacy method into the procedure of recommendation. We design two differentially private recommender algorithms with sampling, named Differentially Private Item Based Recommendation with sampling (DP-IR for short) and Differentially Private User Based Recommendation with sampling (DP-UR for short). Both algorithms are based on the exponential mechanism with a carefully designed quality function. Theoretical analyses on privacy of these algorithms are presented. We also investigate the accuracy of the proposed method and give theoretical results. Experiments are performed on real datasets to verify our methods.
Persistent Identifierhttp://hdl.handle.net/10722/232187
ISBN

 

DC FieldValueLanguage
dc.contributor.authorZhu, X-
dc.contributor.authorSun, Y-
dc.date.accessioned2016-09-20T05:28:19Z-
dc.date.available2016-09-20T05:28:19Z-
dc.date.issued2016-
dc.identifier.citationThe 6th ACM International Workshop on Security and Privacy Analytics (IWSPA 2016), New Orleans, LA., 9-11 March 2016. In Conference Proceedings, 2016, p. 9-16-
dc.identifier.isbn978-1-4503-4077-9-
dc.identifier.urihttp://hdl.handle.net/10722/232187-
dc.description.abstractCollaborative filtering plays an essential role in a recommender system, which recommends a list of items to a user by learning behavior patterns from user rating matrix. However, if an attacker has some auxiliary knowledge about a user purchase history, he/she can infer more information about this user. This brings great threats to user privacy. Some methods adopt differential privacy algorithms in collaborative filtering by adding noises to a rating matrix. Although they provide theoretically private results, the influence on recommendation accuracy are not discussed. In this paper, we solve the privacy problem in recommender system in a different way by applying the differential privacy method into the procedure of recommendation. We design two differentially private recommender algorithms with sampling, named Differentially Private Item Based Recommendation with sampling (DP-IR for short) and Differentially Private User Based Recommendation with sampling (DP-UR for short). Both algorithms are based on the exponential mechanism with a carefully designed quality function. Theoretical analyses on privacy of these algorithms are presented. We also investigate the accuracy of the proposed method and give theoretical results. Experiments are performed on real datasets to verify our methods.-
dc.languageeng-
dc.publisherACM.-
dc.relation.ispartofProceedings of the 2016 ACM on International Workshop on Security And Privacy Analyticsm IWSPA'16-
dc.subjectRecommendation-
dc.subjectCollaborative Filtering-
dc.subjectInference Attack-
dc.subjectDifferential Privacy-
dc.titleDifferential privacy for collaborative filtering recommender algorithm-
dc.typeConference_Paper-
dc.identifier.emailZhu, X: xuezhu26@cs.hku.hk-
dc.description.naturelink_to_OA_fulltext-
dc.identifier.doi10.1145/2875475.2875483-
dc.identifier.scopuseid_2-s2.0-84966670924-
dc.identifier.hkuros267169-
dc.identifier.spage9-
dc.identifier.epage16-
dc.publisher.placeUnited States-
dc.customcontrol.immutablesml 161031-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats