TARU PUBLICATIONS
Journal of Discrete Mathematical Sciences and Cryptography cover
Open Access ·Peer-reviewed·ISSN (Online): 2169-0065·ISSN (Print): 0972-0529

Monthly Journal: Publishes theoretical and applied research in all areas of Discrete Mathematical Sciences, Cryptography, Combinatorics, Elliptic Curves and Information Security.

Issues up to 2022 co-published with and available at:Taylor & Francis Online
submissions@tarupublications.com
Open Access Research Article

Secure SVM training using privacy preserving isomorphic encryption algorithm

, * , , , ,

* Corresponding author · click or hover a name for details

pp. 1497–1504Vol. 28Issue 5-AAugust 2025DOI: 10.47974/JDMSC-2148 Crossmark XML
Received:
05 Nov 2024
Published Online:
30 Aug 2025
Article type:
Research Article
Language:
EN
Article no.:
JDMSC-2148
Pages:
1497–1504

Abstract

Machine Learning algorithm such as Support Vector Machine (/SVM) are identified to be significant for prediting the variables related to the pre-defined output in real world applications. This machine learning model when imposed over the encrypted data is highly indispensable for protecting data and model information against malicious attackers. These malicious adversaries launch the attack over the data either during the phase of prediction or training. Torus-based Fast Fully Homomorphic Encryption (TFFHE) scheme is identified to facilitate potential evaluations of encrypted data related to real numbers, this merits of this TFFHE motivated the option of implementing a privacy preserving machining algorithm which can be utilized during the process of training. In this paper, Secure SVM Training using Privacy Preserving Homomorphic Encryption Algorithm using TFFHE is proposed for preventing inefficient operations and numeric stability during the phase of training in the encrypted domain. This TFFHE is proposed based on the improvement of GSW and its associated ring variants. With respect to real world datasets, this TFFHE-based SVM confirmed better performance on par with the state of the art FHE and logistic regression classifiers used for comparison. This study to the best of the knowledge is one of the few practical algorithms which could be used for SVM model training with the integration of FHE.

Keywords

Subject Classifications

Primary 68P25Secondary 68P27

References

[1] C. Song and X. Shi, “ReActHE: A homomorphic encryption friendly deep neural network for privacy-preserving biomedical prediction,” Smart Health, vol. 32, pp. 100469 (2024).
[2] A. M. Ibrahim, M. Farouk, and M. W. Fakhr, “Privacy preserving image retrieval using multi-key random projection encryption and machine learning decryption,” J. Adv. Res. Appl. Sci. Eng. Technol., vol. 42, no. 2, pp. 155–174 (2024).
[3] A. Hangan, D. Lazea, and T. Cioara, “Privacy preserving anomaly detection on homomorphic encrypted data from IoT sensors,” arXiv preprint, arXiv:2403.09322 (2024).
[4] U. Mehta, J. Vekariya, M. Mehta, H. Kaur, and Y. Kumar, “A review of privacy-preserving machine learning algorithms and systems,” Appl. Data Sci. Smart Syst., pp. 220–225 (2025).
[5] M. K. Morampudi, N. Gonthina, S. Bojjagani, N. K. Sharma, and D. Veeraiah, “Reliable and privacy-preserving multi-instance iris verification using Paillier homomorphic encryption and one-digit checksum,” Signal, Image Video Process., vol. 18, no. 4, pp. 3723–3735 (2024).
[6] B. Mi, J. Zhou, D. Huang, and Y. Weng, “Privacy-preserving data processing method for IoV based on homomorphic conjugacy search problem,” IEEE Trans. Intell. Transp. Syst. (2024).
[7] H. Lee, “Blind evaluation framework for fully homomorphic encryption and privacy-preserving machine learning,” arXiv preprint (2024).
[8] H. Shin, J. Choi, D. Lee, K. Kim, and Y. Lee, “Fully homomorphic training and inference on binary decision tree and random forest,” Cryptology ePrint Archive (2024).
[9] J. Yan, Z. Zhang, Y. Wang, H. Li, X. Liu, and X. Zhang, “Verifiable and privacy-preserving online diagnosis based on multiclass SVM and CKKS leveled homomorphic encryption,” in Proc. 27th Int. Conf. Comput. Supported Cooperative Work Des. (CSCWD), pp. 2794–2799 (May 2024).
[10] A. Falcetta and M. Roveri, “EVAD: Encrypted vibrational anomaly detection with homomorphic encryption,” Neural Comput. Appl., vol. 36, no. 13, pp. 7359–7372 (2024).
[11] C. Guo, Y. Liu, H. Zhang, L. Wang, and J. Chen, “A novel biometric authentication scheme with privacy protection based on SVM and ZKP,” Comput. Secur., pp. 103995 (2024).
[12] R. Wu, B. Wang, and Z. Zhao, “Privacy-preserving medical diagnosis system with Gaussian kernel-based support vector machine,” Peer-to-Peer Netw. Appl., pp. 1–16 (2024).
[13] S. Park, J. Byun, J. Lee, J. H. Cheon, and J. Lee, “HE-friendly algorithm for privacy-preserving SVM training,” IEEE Access, vol. 8, pp. 57414–57425 (2020).
[14] X. Hu, Z. Li, Z. Wang, and X. Lu, “ALT: Area-efficient and low-latency FPGA design for torus fully homomorphic encryption,” IEEE Trans. Very Large Scale Integr. (VLSI) Syst. (2024).
[15] B. Wei and X. Lu, “Improved homomorphic evaluation for hash function based on TFHE,” Cybersecurity, vol. 7, no. 1, pp. 14 (2024).

Views: 154Downloads: 7Citations: 0