Mean estimation over numeric data with personalized local differential privacy

Abstract

The fast development of the Internet and mobile devices results in a crowdsensing business model, where individuals (users) are willing to contribute their data to help the institution (data collector) analyze and release useful information. However, the reveal of personal data will bring huge privacy threats to users, which will impede the wide application of the crowdsensing model. To settle the problem, the definition of local differential privacy (LDP) is proposed. Afterwards, to respond to the varied privacy preference of users, researchers propose a new model, i.e., personalized local differential privacy (PLDP), which allow users to specify their own privacy parameters. In this paper, we focus on a basic task of calculating the mean value over a single numeric attribute with PLDP. Based on the previous schemes for mean estimation under LDP, we employ PLDP model to design novel schemes (LAP, DCP, PWP) to provide personalized privacy for each user. We then theoretically analysis the worst-case variance of three proposed schemes and conduct experiments on synthetic and real datasets to evaluate the performance of three methods. The theoretical and experimental results show the optimality of PWP in the low privacy regime and a slight advantage of DCP in the high privacy regime.

Graphical abstract

Keywords

personalized local differential privacy / mean estimation / crowdsensing model

References

[1] KasiviswanathanS P, LeeH K, NissimK, RaskhodnikovaS. What can we learn privately?. Siam Journal on Computing, 2008, 40( 3): 793– 826

[2] Dwork C. Differential privacy. In: Proceedings of International Conference on Automata, Languages and Programming. 2006, 1-12

[3] DuchiJ C, JordanM I, WainwrightM J. Minimax optimal procedures for locally private estimation. Journal of the American Statistical Association, 2018, 113( 521): 182– 201

[4] Wang N, Xiao X, Yang Y, Zhao J, Hui S, Shin H, Shin J, Yu G. Collecting and Analyzing Multidimensional Data with Local Differential Privacy. In: Proceedings of the 35th IEEE Annual International Conference on Data Engineering. 2019, 638-649

[5] Chen R, Li H, Qin A K, Kasiviswanathan S P, Jin H. Private spatial data aggregation in the local setting. In: Proceedings of the 32nd IEEE International Conference on Data Engineering. 2016, 289-300

[6] Dwork C, McSherry F, Nissim K, Smith A. Calibrating noise to sensitivity in private data analysis. In: Proceedings of the 3rd Theory of Cryptography Conference. 2006, 265-284

[7] LiuY, ZhaoQ. E-Voting scheme using secret sharing and K-anonymity. World Wide Web: Internet and Web Information Systems, 2019, 22( 4): 1657– 1667

[8] XuC, RenJ, ZhangD, ZhangY. Distilling at the edge: a local differential privacy obfuscation framework for IoT data analytics. IEEE Communications Magazine, 2018, 56( 8): 20– 25

[9] ZhangY, HuangH, YangL, XiangY, LiM. Serious challenges and potential solutions for the industrial Internet of Things with edge intelligence. IEEE Network, 2020, 33( 5): 41– 45

[10] KuangB, FuA, YuS, YangG, SuM, ZhangY. ESDRA: an efficient and secure distributed remote attestation scheme for IoT swarms. IEEE Internet of Things Journal, 2019, 6( 5): 8372– 8383

[11] LiN, QardajiW, DongS, CaoJ. Privbasis: frequent itemset mining with differential privacy. Proceedings of the VLDB Endowment, 2012, 5( 11): 1340– 1351

[12] SuS, XuS, ChengX, LiZ, YangF. Differentially private frequent itemset mining via transaction splitting. IEEE Transactions on Knowledge Data Engineering, 2015, 27( 7): 1875– 1891

[13] Zhu Y, Zhang Y, Li X, Yan H, Li J. Improved collusion-resisting secure nearest neighbor query over encrypted data in cloud. Concurrency and Computation Practice and Eperience, 2019, 31(21): e4681

[14] Zhu Y, Li X. Privacy-preserving k-means clustering with local synchronization in peer-to-peer networks. Peer-to-Peer Networking and Applications, 2020, 13(6): 2272−2284

[15] SarwateA D, ChaudhuriK. Signal processing and machine learning with differential privacy: algorithms and challenges for continuous data. IEEE Signal Processing Magazine, 2013, 30( 5): 86– 94

[16] Ji Z, Lipton Z C, Elkan C. Differential privacy and machine learning: a survey and review. 2014, arXiv preprint, arXiv: 1412.7584

[17] ZhangY, XiaoX, YangL, XiangY, ZhongS. Secure and efficient outsourcing of PCA-dased face recognition. IEEE Transactions on Information Forensics and Security, 2020, 15( 1): 1683– 1695

[18] SongJ, LiuY, ShaoJ, TangC. A dynamic membership data aggregation (DMDA) protocol for smart grid. IEEE Systems Journal, 2020, 14( 1): 900– 908

[19] Chen J, Liu G, Liu Y. Lightweight privacy-preserving raw data publishing scheme. IEEE Transactions on Emerging Topics in Computing, 2020,

[20] Fu A, Yu S, Zhang Y, Wang H, Huang C. NPP: a new privacy-aware public auditing scheme for cloud data sharing with group users. IEEE Transactions on Big Data, 2017,

[21] Erlingsson Ú Pihur V, Korolova A. Rappor: randomized aggregatable privacy-preserving ordinal response. In: Proceedings of ACM Sigsac Conference on Computer and Communications Security. 2014, 1054-1067

[22] Kairouz P, Bonawitz K, Ramage D. Discrete distribution estimation under local privacy. In: Proceedings of International Conference on Machine learning. 2016, 2436−2444

[23] YeM, BargA. Optimal schemes for discrete distribution estimation under locally differential privacy. IEEE Transactions on Information Theory, 2018, 64( 8): 5662– 5676

[24] Qin Z, Yang Y, Yu T, Kjalil I, Xiao X, Ren K. Heavy hitter estimation over set-valued data with local differential privacy. In: Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security. 2016, 192-203

[25] Wang T, Blocki J, Li N, Jha S. Locally differentially private protocols for frequency estimation. In: Proceedings of the 26th USENIX Security Symposium. 2017, 729-745

[26] Ye Q, Hu H, Meng X, Zheng H. PrivKV: key-value data collection with local differential privacy. In: Proceedings of IEEE Symposium on Security and Privacy. 2019, 294-308

[27] Xue Q, Zhu Y, Wang J. Joint distribution estimation and naive bayes classification under local differential privacy. IEEE Transactions on Emerging Topics in Computing, 2019,

[28] Xue Q, Zhu Y, Wang J, Li X. Distributed set intersection and union with local differential privacy, In: Proceedings of IEEE International Conference on Parallel & Distributed Systems. 2017, 198-205

[29] Xue Q, Zhu Y, Wang J, Li X, Zhang J. Locally differentially private distributed algorithms for set intersection and union. Science China Information Sciences, 2021, 64: 219101

[30] WarnerS L. Randomized response: a survey technique for eliminating evasive answer bias. Journal of the American Statistical Association, 1965, 60( 309): 63– 66

[31] Jorgensen Z, Yu T, Cormode G. Conservative or liberal? Personalized differential privacy. In: Proceedings of the 32nd IEEE International Conference on Data Engineering. 2016, 1023-1034

[32] Wang S, Huang L, Tian M, Yang W, Xu H, Guo H. Personalized privacy-preserving data aggregation for histogram estimation. In: Proceedings of 2015 IEEE Global Communications Conference. 2015, 1-6

[33] Ye Y, Zhang M, Feng D, Li H, Chi J. Multiple privacy regimes mechanism for local differential privacy. In: Proceedings of International Conference on Database Systems for Advanced Applications. 2019, 247-263

[34] Murakami T, Kawamoto Y. Utility-optimized local differential privacy mechanisms for distribution estimation. In: Proceedings of the 28th USENIX Security Symposium. 2019, 1877-1894

RIGHTS & PERMISSIONS

Higher Education Press