基于差分隐私和SVD++的协同过滤算法
|
摘要
协同过滤技术在推荐系统的实现中具有广泛的应用,协同过滤以用户对商品项目的评价分数为依据,而这些评价有可能反映用户某些不欲为人知的喜好特点,因此,对具备隐私保护能力的协同过滤模型的研究引起了普遍的关注. SVD++是当前最为常用的协同过滤模型之一,差分隐私模型则是近十年来隐私保护理论最重要的研究进展之一,将两者相结合提出3种基于差分隐私和SVD++的协同过滤模型:基于梯度扰动的SVD++隐私保护模型、基于目标函数扰动的SVD++隐私保护模型和基于输出结果扰动的隐私保护模型.理论分析和实验结果显示,所提出的算法不仅能为用户的隐私安全提供可靠的保障,而且还可保持较高的预测准确度.
Collaborative filtering(CF), as a technique that automatically predicts the interest of an user by collecting rating information from other similar users or items, has been widely deployed in various recommendation systems. However,CF prediction is based on the users' historical ratings, indicating that it may reflect some of the users' private preferences.Consequently, enhancing CF model with privacy preservation guarantee has attracted much research attention. In this paper, we propose three privacy preserving collaborative filtering algorithms: DPSS++, DPSAObj++ and DPSAOut++.All the algorithms are based on SVD++, one of the most used CF algorithms, and differential privacy model, one the most important advance in the area of privacy preserving in the last decade. Our analysis shows that the proposed algorithms not only can provide reliable guarantee in terms of privacy preserving, but also keep high prediction accuracy.
Collaborative filtering(CF), as a technique that automatically predicts the interest of an user by collecting rating information from other similar users or items, has been widely deployed in various recommendation systems. However,CF prediction is based on the users' historical ratings, indicating that it may reflect some of the users' private preferences.Consequently, enhancing CF model with privacy preservation guarantee has attracted much research attention. In this paper, we propose three privacy preserving collaborative filtering algorithms: DPSS++, DPSAObj++ and DPSAOut++.All the algorithms are based on SVD++, one of the most used CF algorithms, and differential privacy model, one the most important advance in the area of privacy preserving in the last decade. Our analysis shows that the proposed algorithms not only can provide reliable guarantee in terms of privacy preserving, but also keep high prediction accuracy.
引文
[1] Ricci F, Rokach L, Shapira B. Recommender systems handbook[M]. Berlin:Springer, 2011:1-35.
[2] Linden G, Smith B, York J. Amazon.com recommendations:Item-to-item collaborative filtering[J]. IEEE Internet Computing, 2003, 7(1):76-80.
[3] Zhou Yun-Hong, Wilkinson D, Schreiber R, et al.Large-scale parallel collaborative filtering for the netflix prize[C]. Proc of the 2008 Int Conf on Algorithmic Applications in Management. Binlin:Springer, 2008:337-348.
[4] Narayanan A, Shmatikov V. Robust de-anonymization of large sparse datasets[C]. Proc of the 2008 IEEE Symposium on Security and Privacy. Washington:IEEE,2008:111-125.
[5] Dwork C. Differential privacy[C]. Proc of the 33rd Int Colloquium on Automata, Languages and Programming.Binlin:Springer, 2006:1-12.
[6] Furey T S, Cristianini N, Duffy N, et al. Support vector machine classification and validation of cancer tissue samples using microarray expression data[J].Bioinformatics, 2000, 16(10):906-914.
[7] Dwork C, Roth A. The algorithmic foundations of differential privacy[J]. Foundations&Trends in Theoretical Computer Science, 2014, 9(3/4):211-407.
[8] Mcsherry F, Mironov I. Differentially private recommender systems:Building privacy into the net[C]. Proc of the 2009 ACM SIGKDD Int Conf on Knowledge Discovery and Data Mining. New York:ACM, 2009:627-636.
[9] Hardt M, Roth A. Beating randomized response on incoherent matrices[C]. Proc of the Annual Acm Symposium on the Theory of computing. New York:ACM, 2012:1255-1268.
[10]鲜征征,李启良.差分隐私在推荐系统中的应用研究[J].计算机应用研究, 2016, 33(5):1549-1553.(Xian Z Z, Li Q L. Research on application of differential privacy in recommender system[J]. Application Research of Computers, 2016, 33(5):1549-1553.)
[11] Hua Jing-yu, Xia Chang, Zhong Sheng. Differential private matrix factorization[J]. Proc of the 24th Int Conf on Artificial Intelligence. Palo Alto:AAAI Press, 2015:1763-1770.
[12] Yan Shen, Pan Shi-Ran, Zhu Wen-Tao, et al. DynaEgo:Privacy-reserving collaborative filtering recommender system based on social-Aware differential privacy[C].Proc of the Int Conf on Information and Communications Security. Binlin:Springer, 2016:347-357.
[13] Friedman A, Berkovsky S, Kaafar M A. A differential privacy framework for matrix factorization recommender systems[J]. User Modeling and User-Adapted Interaction,2016, 26(5):1-34.
[14] Koren Y. Factorization meets the neighborhood:A multifaceted collaborative filtering model[C]. Proc of the2008 Acm Sigkdd Int Conf on Knowledg Discovery and Data Mining. New York:ACM, 2008:426-434.
[15] Koren Y, Bell R, Volinsky C. Matrix factorization techniques for recommender systems[J]. Computer, 2009,42(8):30-37.
[16] Dwork C. Differential privacy:A survey of results[C].Proc of the 5th Int Conf on Theory and Applications of Models of Computation. Binlin:Springer, 2008:1-19.
[17] McSherry F. Privacy integrated queries:An extensible platform for privacy-preserving data analysis[C]. Proc of the 2009 ACM SIGMOD Int Conf on Management of Data(SIGMOD). New York:ACM, 2009:19-30.
[18] Nissim K, Raskhodnikova S, Smith A. Smooth sensitivity and sampling in private data analysis[C]. Proc of the 39th Annual ACM Symposium on Theory of Computing. New York:ACM, 2007:75-84.
[19] Dwork C, McSherry F, Nissim K, et al. Calibrating noise to sensitivity in private data analysis[C]. Proc of the3th Theory of Cryptography Conference(TCC). Binlin:Springer, 2006:363-385.
[20]熊平,朱天清,王晓峰.差分隐私保护及其应用[J].计算机学报, 2014, 37(1):101-122.(Xiong P, Zhu T Q, Wang X F. A survey on differential privacy and applications[J]. Chinese J of Computers,2014, 37(1):101-122.)
[21] McSherry F. Privacy integrated queries:an extensible platform for privacy-preserving data analysis[C]. Proc of the 2009 ACM SIGMOD Int Conf on Management of Data(SIGMOD). New York:ACM, 2009:19-30.
[22] Chaudhuri K, Monteleoni C, Sarwate A. Differentially private empirical risk minimization[J]. J of Machine Learning Research, 2011, 12(2):1069-1109.
[23] Su Kai, Ma Liang-Li, Xiao Bin, et al. Web service QoS prediction by neighbor information combined non-negative matrix factorization[J]. J of Intelligent&Fuzzy Systems, 2016, 30(6):3593-3604.
[2] Linden G, Smith B, York J. Amazon.com recommendations:Item-to-item collaborative filtering[J]. IEEE Internet Computing, 2003, 7(1):76-80.
[3] Zhou Yun-Hong, Wilkinson D, Schreiber R, et al.Large-scale parallel collaborative filtering for the netflix prize[C]. Proc of the 2008 Int Conf on Algorithmic Applications in Management. Binlin:Springer, 2008:337-348.
[4] Narayanan A, Shmatikov V. Robust de-anonymization of large sparse datasets[C]. Proc of the 2008 IEEE Symposium on Security and Privacy. Washington:IEEE,2008:111-125.
[5] Dwork C. Differential privacy[C]. Proc of the 33rd Int Colloquium on Automata, Languages and Programming.Binlin:Springer, 2006:1-12.
[6] Furey T S, Cristianini N, Duffy N, et al. Support vector machine classification and validation of cancer tissue samples using microarray expression data[J].Bioinformatics, 2000, 16(10):906-914.
[7] Dwork C, Roth A. The algorithmic foundations of differential privacy[J]. Foundations&Trends in Theoretical Computer Science, 2014, 9(3/4):211-407.
[8] Mcsherry F, Mironov I. Differentially private recommender systems:Building privacy into the net[C]. Proc of the 2009 ACM SIGKDD Int Conf on Knowledge Discovery and Data Mining. New York:ACM, 2009:627-636.
[9] Hardt M, Roth A. Beating randomized response on incoherent matrices[C]. Proc of the Annual Acm Symposium on the Theory of computing. New York:ACM, 2012:1255-1268.
[10]鲜征征,李启良.差分隐私在推荐系统中的应用研究[J].计算机应用研究, 2016, 33(5):1549-1553.(Xian Z Z, Li Q L. Research on application of differential privacy in recommender system[J]. Application Research of Computers, 2016, 33(5):1549-1553.)
[11] Hua Jing-yu, Xia Chang, Zhong Sheng. Differential private matrix factorization[J]. Proc of the 24th Int Conf on Artificial Intelligence. Palo Alto:AAAI Press, 2015:1763-1770.
[12] Yan Shen, Pan Shi-Ran, Zhu Wen-Tao, et al. DynaEgo:Privacy-reserving collaborative filtering recommender system based on social-Aware differential privacy[C].Proc of the Int Conf on Information and Communications Security. Binlin:Springer, 2016:347-357.
[13] Friedman A, Berkovsky S, Kaafar M A. A differential privacy framework for matrix factorization recommender systems[J]. User Modeling and User-Adapted Interaction,2016, 26(5):1-34.
[14] Koren Y. Factorization meets the neighborhood:A multifaceted collaborative filtering model[C]. Proc of the2008 Acm Sigkdd Int Conf on Knowledg Discovery and Data Mining. New York:ACM, 2008:426-434.
[15] Koren Y, Bell R, Volinsky C. Matrix factorization techniques for recommender systems[J]. Computer, 2009,42(8):30-37.
[16] Dwork C. Differential privacy:A survey of results[C].Proc of the 5th Int Conf on Theory and Applications of Models of Computation. Binlin:Springer, 2008:1-19.
[17] McSherry F. Privacy integrated queries:An extensible platform for privacy-preserving data analysis[C]. Proc of the 2009 ACM SIGMOD Int Conf on Management of Data(SIGMOD). New York:ACM, 2009:19-30.
[18] Nissim K, Raskhodnikova S, Smith A. Smooth sensitivity and sampling in private data analysis[C]. Proc of the 39th Annual ACM Symposium on Theory of Computing. New York:ACM, 2007:75-84.
[19] Dwork C, McSherry F, Nissim K, et al. Calibrating noise to sensitivity in private data analysis[C]. Proc of the3th Theory of Cryptography Conference(TCC). Binlin:Springer, 2006:363-385.
[20]熊平,朱天清,王晓峰.差分隐私保护及其应用[J].计算机学报, 2014, 37(1):101-122.(Xiong P, Zhu T Q, Wang X F. A survey on differential privacy and applications[J]. Chinese J of Computers,2014, 37(1):101-122.)
[21] McSherry F. Privacy integrated queries:an extensible platform for privacy-preserving data analysis[C]. Proc of the 2009 ACM SIGMOD Int Conf on Management of Data(SIGMOD). New York:ACM, 2009:19-30.
[22] Chaudhuri K, Monteleoni C, Sarwate A. Differentially private empirical risk minimization[J]. J of Machine Learning Research, 2011, 12(2):1069-1109.
[23] Su Kai, Ma Liang-Li, Xiao Bin, et al. Web service QoS prediction by neighbor information combined non-negative matrix factorization[J]. J of Intelligent&Fuzzy Systems, 2016, 30(6):3593-3604.