How to Profile Privacy-Conscious Users in Recommender Systems

Matrix factorization is a popular method to build a recommender system. In such a system, existing users and items are associated to a low-dimension vector called a profile. The profiles of a user and of an item can be combined (via inner product) to predict the rating that the user would get on the item. One important issue of such a system is the so-called cold-start problem: how to allow a user to learn her profile, so that she can then get accurate recommendations? While a profile can be computed if the user is willing to rate well-chosen items and/or provide supplemental attributes or demographics (such as gender), revealing this additional information is known to allow the analyst of the recommender system to infer many more personal sensitive information. We design a protocol to allow privacy-conscious users to benefit from matrix-factorization-based recommender systems while preserving their privacy. More precisely, our protocol enables a user to learn her profile, and from that to predict ratings without the user revealing any personal information. The protocol is secure in the standard model against semi-honest adversaries.

[1]  Vitaly Shmatikov,et al.  2011 IEEE Symposium on Security and Privacy “You Might Also Like:” Privacy Risks of Collaborative Filtering , 2022 .

[2]  Silvio Micali,et al.  Probabilistic Encryption , 1984, J. Comput. Syst. Sci..

[3]  T. Graepel,et al.  Private traits and attributes are predictable from digital records of human behavior , 2013, Proceedings of the National Academy of Sciences.

[4]  Sean M. McNee,et al.  Getting to know you: learning new user preferences in recommender systems , 2002, IUI '02.

[5]  Moni Naor,et al.  Efficient oblivious transfer protocols , 2001, SODA '01.

[6]  Stratis Ioannidis,et al.  BlurMe: inferring and obfuscating user gender based on ratings , 2012, RecSys.

[7]  James H. Davenport,et al.  P-adic reconstruction of rational numbers , 1982, SIGS.

[8]  Gediminas Adomavicius,et al.  Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions , 2005, IEEE Transactions on Knowledge and Data Engineering.

[9]  Stratis Ioannidis,et al.  Privacy tradeoffs in predictive analytics , 2014, SIGMETRICS '14.

[10]  Rafail Ostrovsky,et al.  A Survey of Single-Database Private Information Retrieval: Techniques and Applications , 2007, Public Key Cryptography.

[11]  Vitaly Shmatikov,et al.  Robust De-anonymization of Large Sparse Datasets , 2008, 2008 IEEE Symposium on Security and Privacy (sp 2008).

[12]  Jacques Stern,et al.  CryptoComputing with Rationals , 2002, Financial Cryptography.

[13]  David M. Pennock,et al.  Categories and Subject Descriptors , 2001 .

[14]  Yehuda Koren,et al.  Factorization meets the neighborhood: a multifaceted collaborative filtering model , 2008, KDD.

[15]  Murphy J. Stephen,et al.  You Might Also Like , 2014 .

[16]  Stratis Ioannidis,et al.  Recommending with an agenda: active learning of private attributes using matrix factorization , 2013, RecSys '14.

[17]  Patrick Seemann,et al.  Matrix Factorization Techniques for Recommender Systems , 2014 .

[18]  Nina Taft,et al.  How to hide the elephant- or the donkey- in the room: Practical privacy against statistical inference for large data , 2013, 2013 IEEE Global Conference on Signal and Information Processing.

[19]  Andrea Montanari,et al.  Learning low rank matrices from O(n) entries , 2008, 2008 46th Annual Allerton Conference on Communication, Control, and Computing.

[20]  Emmanuel J. Candès,et al.  Exact Matrix Completion via Convex Optimization , 2008, Found. Comput. Math..

[21]  Stratis Ioannidis,et al.  Privacy-preserving matrix factorization , 2013, CCS.

[22]  Pascal Paillier,et al.  Public-Key Cryptosystems Based on Composite Degree Residuosity Classes , 1999, EUROCRYPT.