Abstract
AbstractNowadays, computing on encrypted data seems to be more practical than a few years ago, thanks to the emergence of new Homomorphic Encryption schemes. In this paper, an algorithm based on Homomorphic Encryption for Arithmetic of Approximate Numbers (Cheon et al., in: Takagi, Peyrin (eds) Advances in cryptology—ASIACRYPT 2017, Springer, Cham, pp 409–437, 2017) (HEAAN, or also CKKS) scheme, that is able to perform a secure k-means algorithm which processes encrypted data, has been studied and presented. The performance of the classifier running on encrypted data has been evaluated using a standard k-means algorithm that works on plain data as a supervised structure, since the results are obtained by approximated computations. The main point of this paper is to take existent theoretical techniques (for example approximations of $$\text {sgn}(x)$$
sgn
(
x
)
), to use them and to observe if they are valid in practical applications. The output of the algorithm is a set of k encrypted masks that can be applied to the original dataset in order to obtain different clusters. The setting is a standard client–server one. The workload is heavily server-centric, as the client only has to execute a light masking algorithm at the end of each iteration, which, excluding the decryption, is faster than a plain k-means iteration; the main disadvantage concerns the accuracy of the results. Experiments show that the algorithm can be executed fairly quickly: the execution time of the training phase is on the order of seconds, while classification is on the order of tenths of a second.
Funder
Università degli Studi di Milano - Bicocca
Publisher
Springer Science and Business Media LLC
Subject
Computer Networks and Communications,Safety, Risk, Reliability and Quality,Information Systems,Software
Reference39 articles.
1. Bourse, F., Minelli, M., Minihold, M., Paillier, P.: Fast homomorphic evaluation of deep discretized neural networks. In: Shacham, H., Boldyreva, A. (eds.) Advances in Cryptology—CRYPTO 2018, pp. 483–512. Springer, Cham (2018)
2. Brakerski, Z.: Fully homomorphic encryption without modulus switching from classical gapsvp. In: Proceedings of the 32nd Annual Cryptology Conference on Advances in Cryptology—CRYPTO 2012—Volume 7417, pp. 868-886. Springer, Berlin (2012)
3. Brakerski, Z., Gentry, C., Vaikuntanathan, V.: Fully homomorphic encryption without bootstrapping. Cryptology ePrint Archive, Paper 2011/277. (2011) https://eprint.iacr.org/2011/277
4. Bunn, P. ,Ostrovsky, R.: Secure two-party k-means clustering. In Proceedings of the 14th ACM Conference on Computer and Communications Security, CCS ’07, pp 486-497. Association for Computing Machinery, New York (2007)
5. Catak, F.O., Aydin, I., Elezaj, O., Yildirim-Yayilgan, S.: Practical implementation of privacy preserving clustering methods using a partially homomorphic encryption algorithm. Electronics 9(2), 1 (2020)
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. FHE-Based Secure Image Processing Framework with Integrated Key Management System;The Fifth Workshop on Intelligent Cross-Data Analysis and Retrieval;2024-06-10
2. Homomorphic Encryption;Advances to Homomorphic and Searchable Encryption;2023