{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,8,30]],"date-time":"2024-08-30T23:52:18Z","timestamp":1725061938341},"reference-count":62,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2023,10,12]],"date-time":"2023-10-12T00:00:00Z","timestamp":1697068800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Big Data"],"abstract":"State-of-the-art recommender systems produce high-quality recommendations to support users in finding relevant content. However, through the utilization of users' data for generating recommendations, recommender systems threaten users' privacy. To alleviate this threat, often, differential privacy is used to protect users' data via adding random noise. This, however, leads to a substantial drop in recommendation quality. Therefore, several approaches aim to improve this trade-off between accuracy and user privacy. In this work, we first overview threats to user privacy in recommender systems, followed by a brief introduction to the differential privacy framework that can protect users' privacy. Subsequently, we review recommendation approaches that apply differential privacy, and we highlight research that improves the trade-off between recommendation quality and user privacy. Finally, we discuss open issues, e.g., considering the relation between privacy and fairness, and the users' different needs for privacy. With this review, we hope to provide other researchers an overview of the ways in which differential privacy has been applied to state-of-the-art collaborative filtering recommender systems.<\/jats:p>","DOI":"10.3389\/fdata.2023.1249997","type":"journal-article","created":{"date-parts":[[2023,10,15]],"date-time":"2023-10-15T01:31:26Z","timestamp":1697333486000},"update-policy":"http:\/\/dx.doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":4,"title":["Differential privacy in collaborative filtering recommender systems: a review"],"prefix":"10.3389","volume":"6","author":[{"given":"Peter","family":"M\u00fcllner","sequence":"first","affiliation":[]},{"given":"Elisabeth","family":"Lex","sequence":"additional","affiliation":[]},{"given":"Markus","family":"Schedl","sequence":"additional","affiliation":[]},{"given":"Dominik","family":"Kowald","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2023,10,12]]},"reference":[{"key":"B1","volume-title":"Trade-offs between fairness, interpretability, and privacy in machine learning","author":"Agarwal","year":"2020"},{"key":"B2","doi-asserted-by":"publisher","first-page":"302","DOI":"10.1137\/060673096","article-title":"The fast Johnson\u2013Lindenstrauss transform and approximate nearest neighbors","volume":"39","author":"Ailon","year":"2009","journal-title":"SIAM J. Comput."},{"key":"B3","first-page":"2938","article-title":"\u201cHow to backdoor federated learning,\u201d","volume-title":"International Conference on Artificial Intelligence and Statistics","author":"Bagdasaryan","year":"2020"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3343038","article-title":"A survey on privacy in social media: identification, mitigation, and applications","volume":"1","author":"Beigi","year":"2020","journal-title":"ACM Trans. Data Sci."},{"key":"B5","doi-asserted-by":"publisher","first-page":"5033","DOI":"10.1016\/j.eswa.2011.11.037","article-title":"The impact of data obfuscation on the accuracy of collaborative filtering","volume":"39","author":"Berkovsky","year":"2012","journal-title":"Expert Syst. Appl."},{"key":"B6","doi-asserted-by":"publisher","first-page":"1085","DOI":"10.1142\/S0218194013500320","article-title":"A survey of privacy-preserving collaborative filtering schemes","volume":"23","author":"Bilge","year":"2013","journal-title":"Int. J. Softw. Eng. Knowledge Eng."},{"key":"B7","doi-asserted-by":"crossref","first-page":"410","DOI":"10.1109\/FOCS.2012.67","article-title":"\u201cThe Johnson-Lindenstrauss transform itself preserves differential privacy,\u201d","volume-title":"2012 IEEE 53rd Annual Symposium on Foundations of Computer Science","author":"Blocki","year":"2012"},{"key":"B8","first-page":"231","article-title":"\u201c\u201cYou might also like:\u201d privacy risks of collaborative filtering,\u201d","volume-title":"Proc. of S&P'11","author":"Calandrino","year":"2011"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3501812","article-title":"Efficient federated matrix factorization against inference attacks","volume":"13","author":"Chai","year":"2022","journal-title":"ACM Trans. Intell. Syst. Technol."},{"key":"B10","doi-asserted-by":"crossref","DOI":"10.1145\/3485447.3512192","article-title":"\u201cDifferential private knowledge transfer for privacy-preserving cross-domain recommendation,\u201d","volume-title":"Proc. of ACM WWW'22","author":"Chen","year":"2022"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1455","DOI":"10.1145\/3394138","article-title":"Practical privacy preserving POI recommendation","volume":"11","author":"Chen","year":"2020","journal-title":"ACM Trans. Intell. Syst. Technol."},{"key":"B12","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/978-3-540-79228-4_1","article-title":"\u201cDifferential privacy: a survey of results,\u201d","volume-title":"International Conference on Theory and Applications of Models of Computation","author":"Dwork","year":"2008"},{"key":"B13","first-page":"214","article-title":"\u201cFairness through awareness,\u201d","volume-title":"Proc. of ITCS'12","author":"Dwork","year":"2012"},{"key":"B14","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1561\/0400000042","article-title":"The algorithmic foundations of differential privacy","volume":"9","author":"Dwork","year":"2014","journal-title":"Found. Trends Theoret. Comput. Sci."},{"key":"B15","first-page":"35","article-title":"\u201cPrivacy for all: Ensuring fair and equitable privacy protections,\u201d","volume-title":"Proc. of FAccT'18","author":"Ekstrand","year":"2018"},{"key":"B16","doi-asserted-by":"crossref","first-page":"123","DOI":"10.1007\/978-3-030-17138-4_6","article-title":"\u201cGeneralised differential privacy for text document processing,\u201d","volume-title":"Principles of Security and Trust: 8th International Conference, POST 2019","author":"Fernandes","year":"2019"},{"key":"B17","doi-asserted-by":"publisher","first-page":"425","DOI":"10.1007\/s11257-016-9177-7","article-title":"A differential privacy framework for matrix factorization recommender systems","volume":"26","author":"Friedman","year":"2016","journal-title":"User Model. User Adapt. Interact."},{"key":"B18","doi-asserted-by":"crossref","first-page":"649","DOI":"10.1007\/978-1-4899-7637-6_19","article-title":"\u201cPrivacy aspects of recommender systems,\u201d","volume-title":"Recommender Systems Handbook","author":"Friedman","year":"2015"},{"key":"B19","doi-asserted-by":"crossref","first-page":"2142","DOI":"10.1145\/3477495.3531820","article-title":"\u201cUnlearning protected user attributes in recommendations with adversarial training,\u201d","volume-title":"SIGIR '22: The 45th International ACM SIGIR Conference on Research and Development in Information Retrieval","author":"Ganh\u00f6r","year":"2022"},{"key":"B20","doi-asserted-by":"crossref","first-page":"961","DOI":"10.1145\/3397271.3401053","article-title":"\u201cDPLCF: differentially private local collaborative filtering,\u201d","volume-title":"Proc. of SIGIR'20","author":"Gao","year":"2020"},{"key":"B21","author":"Gentry","year":"2009","journal-title":"A Fully Homomorphic Encryption Scheme (Dissertation)"},{"key":"B22","article-title":"Data leakage via access patterns of sparse features in deep learning-based recommendation systems","author":"Hashemi","year":"2022","journal-title":"arXiv preprint arXiv:2212.06264"},{"key":"B23","doi-asserted-by":"publisher","first-page":"763196","DOI":"10.3389\/fdata.2021.763196","article-title":"Are you willing to self-disclose for science? Effects of privacy awareness and trust in privacy on self-disclosure of personal and health data in online scientific studies\u2013an experimental study","volume":"4","author":"Herbert","year":"2021","journal-title":"Front. Big Data"},{"key":"B24","doi-asserted-by":"publisher","first-page":"102746","DOI":"10.1016\/j.cose.2022.102746","article-title":"Latest trends of security and privacy in recommender systems: a comprehensive review and future perspectives","volume":"118","author":"Himeur","year":"2022","journal-title":"Comput. Sec."},{"key":"B25","article-title":"\u201cDifferentially private matrix factorization,\u201d","volume-title":"International Joint Conference on Artificial Intelligence","author":"Hua","year":"2015"},{"key":"B26","doi-asserted-by":"crossref","first-page":"263","DOI":"10.1007\/978-1-4471-4555-4_12","article-title":"\u201cPrivacy in recommender systems,\u201d","volume-title":"Social Media Retrieval","author":"Jeckmans","year":"2013"},{"key":"B27","first-page":"1","article-title":"\u201cFedNCF: federated neural collaborative filtering for privacy-preserving recommender system,\u201d","volume-title":"2022 International Joint Conference on Neural Networks (IJCNN)","author":"Jiang","year":"2022"},{"key":"B28","article-title":"\u201cDo different groups have comparable privacy tradeoffs?\u201d","volume-title":"Workshop on Moving Beyond a \u201cOne-Size Fits All\u201d Approach: Exploring Individual Differences in Privacy, in Conjunction with the ACM CHI Conference on Human Factors in Computing Systems, CHI 2018","author":"Joshaghani","year":"2018"},{"key":"B29","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/2499670","article-title":"Making decisions about privacy: information disclosure in context-aware recommender systems","volume":"3","author":"Knijnenburg","year":"2013","journal-title":"ACM Trans. Interact. Intell. Syst."},{"key":"B30","doi-asserted-by":"publisher","first-page":"5802","DOI":"10.1073\/pnas.1218772110","article-title":"Private traits and attributes are predictable from digital records of human behavior","volume":"110","author":"Kosinski","year":"2013","journal-title":"Proc. Nat. Acad. Sci. U.S.A."},{"key":"B31","doi-asserted-by":"publisher","first-page":"35","DOI":"10.1109\/ACCESS.2016.2600258","article-title":"Enforcing differential privacy for shared collaborative filtering","volume":"5","author":"Li","year":"2016","journal-title":"IEEE Access"},{"key":"B32","doi-asserted-by":"publisher","first-page":"900","DOI":"10.14778\/3503585.3503598","article-title":"Federated matrix factorization with privacy guarantee","volume":"15","author":"Li","year":"2021","journal-title":"Proc. VLDB Endowment"},{"key":"B33","doi-asserted-by":"crossref","first-page":"576","DOI":"10.1007\/978-3-319-55753-3_36","article-title":"\u201cWhen differential privacy meets randomized perturbation: a hybrid approach for privacy-preserving recommender system,\u201d","volume-title":"International Conference on Database Systems for Advanced Applications","author":"Liu","year":"2017"},{"key":"B34","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3501815","article-title":"Federated social recommendation with graph neural network","volume":"13","author":"Liu","year":"2022","journal-title":"ACM Trans. Intell. Syst. Technol."},{"key":"B35","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3555374","article-title":"Decentralized collaborative learning framework for next POI recommendation","volume":"41","author":"Long","year":"2023","journal-title":"ACM Trans. Inf. Syst."},{"key":"B36","first-page":"1291","article-title":"\u201cPrivacy-preserving tensor factorization for collaborative health data analysis,\u201d","volume-title":"Proceedings of the 28th ACM International Conference on Information and Knowledge Management, CIKM '19","author":"Ma","year":"2019"},{"key":"B37","doi-asserted-by":"publisher","first-page":"8512","DOI":"10.1109\/ACCESS.2020.3045700","article-title":"Anonymization techniques for privacy preserving data publishing: a comprehensive survey","volume":"9","author":"Majeed","year":"2020","journal-title":"IEEE Access"},{"key":"B38","first-page":"1273","article-title":"\u201cCommunication-efficient learning of deep networks from decentralized data,\u201d","volume-title":"Artificial Intelligence and Statistics","author":"McMahan","year":"2017"},{"key":"B39","doi-asserted-by":"crossref","DOI":"10.1145\/3450613.3456829","article-title":"\u201cPrivacy as a planned behavior: effects of situational factors on privacy perceptions and plans,\u201d","volume-title":"Proc. of UMAP'21","author":"Mehdy","year":"2021"},{"key":"B40","doi-asserted-by":"crossref","DOI":"10.1609\/aaai.v32i1.11714","article-title":"\u201cPersonalized privacy-preserving social recommendation,\u201d","volume-title":"Proceedings of the AAAI Conference on Artificial Intelligence","author":"Meng","year":"2018"},{"key":"B41","first-page":"342","article-title":"\u201cStronger privacy for federated collaborative filtering with implicit feedback,\u201d","volume-title":"Proceedings of the 15th ACM Conference on Recommender Systems","author":"Minto","year":"2021"},{"key":"B42","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3608481","article-title":"ReuseKNN: neighborhood reuse for differentially-private KNN-based recommendations","volume":"14","author":"M\u00fcllner","year":"2023","journal-title":"ACM Trans. Intell. Syst. Technol."},{"key":"B43","doi-asserted-by":"publisher","first-page":"4151","DOI":"10.1109\/TKDE.2021.3126577","article-title":"Private and utility enhanced recommendations with local differential privacy and Gaussian mixture model","volume":"35","author":"Neera","year":"2023","journal-title":"IEEE Trans. Knowledge Data Eng."},{"key":"B44","doi-asserted-by":"crossref","first-page":"399","DOI":"10.1145\/3523227.3546769","article-title":"\u201cEANA: reducing privacy risk on large-scale recommendation models,\u201d","volume-title":"Proceedings of the 16th ACM Conference on Recommender Systems","author":"Ning","year":"2022"},{"key":"B45","doi-asserted-by":"publisher","first-page":"54","DOI":"10.1109\/4236.968832","article-title":"When being weak is brave: privacy in recommender systems","volume":"5","author":"Ramakrishnan","year":"2001","journal-title":"IEEE Internet Comput."},{"key":"B46","doi-asserted-by":"publisher","first-page":"32","DOI":"10.1016\/j.neucom.2022.01.079","article-title":"A differentially private matrix factorization based on vector perturbation for recommender system","volume":"483","author":"Ran","year":"2022","journal-title":"Neurocomputing"},{"key":"B47","first-page":"1685","article-title":"\u201cCross-network social user embedding with hybrid differential privacy guarantees,\u201d","volume-title":"Proceedings of the 31st ACM International Conference on Information &; Knowledge Management, CIKM '22","author":"Ren","year":"2022"},{"key":"B48","doi-asserted-by":"publisher","first-page":"1770","DOI":"10.1109\/TKDE.2018.2805356","article-title":"Privacy enhanced matrix factorization for recommendation with local differential privacy","volume":"30","author":"Shin","year":"2018","journal-title":"IEEE Trans. Knowledge Data Eng."},{"key":"B49","doi-asserted-by":"crossref","first-page":"188","DOI":"10.1007\/978-3-031-28238-6_13","article-title":"\u201cPrivacy-preserving fair item ranking,\u201d","volume-title":"Advances in Information Retrieval","author":"Sun","year":"2023"},{"key":"B50","doi-asserted-by":"publisher","first-page":"119457","DOI":"10.1016\/j.eswa.2022.119457","article-title":"An improved matrix factorization with local differential privacy based on piecewise mechanism for recommendation systems","volume":"216","author":"Wang","year":"2023","journal-title":"Expert Syst. Appl."},{"key":"B51","doi-asserted-by":"crossref","first-page":"63","DOI":"10.1080\/01621459.1965.10480775","article-title":"Randomized response: a survey technique for eliminating evasive answer bias","volume":"60","author":"Warner","year":"1965","journal-title":"J. Am. Stat. Assoc."},{"key":"B52","doi-asserted-by":"crossref","first-page":"195","DOI":"10.1145\/2365952.2365989","article-title":"\u201cBlurme: inferring and obfuscating user gender based on ratings,\u201d","volume-title":"Proceedings of the Sixth ACM Conference on Recommender Systems","author":"Weinsberg","year":"2012"},{"key":"B53","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3506715","article-title":"FedCTR: Federated native ad CTR prediction with cross-platform user behavior data","volume":"13","author":"Wu","year":"2022","journal-title":"ACM TIST"},{"key":"B54","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3568954","article-title":"On the user behavior leakage from recommender system exposure","volume":"41","author":"Xin","year":"2023","journal-title":"ACM Trans. Inform. Syst."},{"key":"B55","doi-asserted-by":"crossref","first-page":"417","DOI":"10.1109\/Trustcom\/BigDataSE\/ICESS.2017.266","article-title":"\u201cPrivacy preserving collaborative filtering via the Johnson-Lindenstrauss transform,\u201d","volume-title":"2017 IEEE Trustcom\/BigDataSE\/ICESS","author":"Yang","year":"2017"},{"key":"B56","doi-asserted-by":"crossref","first-page":"927","DOI":"10.1145\/3543873.3587577","article-title":"\u201cFairness-aware differentially private collaborative filtering,\u201d","volume-title":"Companion Proceedings of the ACM Web Conference 2023, WWW '23 Companion","author":"Yang","year":"2023"},{"key":"B57","first-page":"325","article-title":"\u201cLearning fair representations,\u201d","volume-title":"Proc. of ICML'13","author":"Zemel","year":"2013"},{"key":"B58","doi-asserted-by":"crossref","first-page":"3002","DOI":"10.1145\/3442381.3449813","article-title":"\u201cGraph embedding for recommendation against attribute inference attacks,\u201d","volume-title":"Proceedings of the Web Conference 2021, WWW '21","author":"Zhang","year":"2021"},{"key":"B59","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2023.3295601","article-title":"Comprehensive privacy analysis on federated recommender system against attribute inference attacks","author":"Zhang","year":"2023","journal-title":"IEEE Trans. Knowledge Data Eng"},{"key":"B60","doi-asserted-by":"crossref","first-page":"29","DOI":"10.1145\/2043932.2043942","article-title":"\u201cShared collaborative filtering,\u201d","volume-title":"Proceedings of the Fifth ACM Conference on Recommender Systems","author":"Zhao","year":"2011"},{"key":"B61","doi-asserted-by":"publisher","first-page":"8836","DOI":"10.1109\/JIOT.2020.3037194","article-title":"Local differential privacy-based federated learning for internet of things","volume":"8","author":"Zhao","year":"2020","journal-title":"IEEE Internet Things J."},{"key":"B62","doi-asserted-by":"crossref","first-page":"752","DOI":"10.1145\/2492517.2492519","article-title":"\u201cDifferential privacy for neighborhood-based collaborative filtering,\u201d","volume-title":"Proc. of IEEE\/ACM ASONAM'13","author":"Zhu","year":"2013"}],"container-title":["Frontiers in Big Data"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdata.2023.1249997\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,10,15]],"date-time":"2023-10-15T01:31:47Z","timestamp":1697333507000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdata.2023.1249997\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,10,12]]},"references-count":62,"alternative-id":["10.3389\/fdata.2023.1249997"],"URL":"https:\/\/doi.org\/10.3389\/fdata.2023.1249997","relation":{},"ISSN":["2624-909X"],"issn-type":[{"value":"2624-909X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,10,12]]}}}