Article Details

Abstract

To address the issues of user privacy leakage and excessive system overhead in recommendation systems, this paper proposes a lightweight privacy-preserving recommendation protocol based on domingo-ferrer homomorphic encryption. The protocol constructs a three-party model consisting of the target user, the recommendation server, and the privacy server. By integrating users’ location information, interest preferences, and social relationships, it achieves personalized location recommendations through multidimensional similarity computation. While ensuring that sensitive data remain undisclosed, the protocol optimizes the encrypted computation process and reduces both communication and computation costs. Experiments conducted on a public check-in dataset demonstrate that the proposed method outperforms existing approaches in terms of recommendation accuracy, operational efficiency, and privacy protection, indicating strong practicality and applicability.

Keywords

privacy protection; homomorphic encryption; location service; recommendation system; similarity computation; lightweight protocol

References

1. [1] PRAMOD D. Privacy-preserving Techniques in Recommender Systems: State-of-the-art Review and Future Research Agenda [J].Data Technologies and Applications，2023，57（1）：32-55.
2. [2] Peng Jing Research on User Privacy Protection Protocol in Recommendation System [D]. Quanzhou: Huaqiao University, 2023
3. [3] PAILLIER P. Public-key Cryptosystems Based on Composite Degree Residuosity Classes [C]//International Conference on the Theory and Applications of Cryptographic Techniques.Prague：Springer，1999：223-238.
4. [4] DOMINGO-FERRER J. A Provably Secure Additive and Multiplicative Privacy Homomorphism [J].Information Security，2002，180（19）：3666-3682.
5. [5] Ma Xindi, Li Hui, Ma Jianfeng, etc Privacy Protection Framework for Lightweight Location Aware Recommendation System [J]. Journal of Computer Science, 2017, 40 (5): 1017-1030
6. [6] CHRONIS C，VARLAMIS I，HIMEUR Y，et al. A Survey on the Use of Federated Learning in Privacy-preserving Recommender Systems [J].IEEE Open Journal of the Computer Society，2024，5：227-247.
7. [7] Pan Yiteng, He Fazhi, Yu Haiping A social recommendation algorithm based on trust relationship implicit similarity [J]. Journal of Computer Science, 2018, 41 (1): 65-81
8. [8] Yang Jun, Wu Juhua, Ai Danxiang A New Method for Multidimensional Information Recommendation Based on Scene Similarity [J]. Journal of Intelligence, 2013, 32 (3): 262-269
9. [9] Guo Xian, Wang Diandong, Feng Tao, etc A verifiable privacy preserving federated learning scheme based on homomorphic encryption [J]. Journal of Electronics and Information Technology, 2025, 47 (4): 1113-1125
10. [10] ARMKNECHT F，STRUFE T. An Efficient Distributed Privacy-preserving Recommendation System [C]//2011 the 10th IFIP Annual Mediterranean AD Hoc Networking Workshop.Favignana Island：IEEE，2011：65-70.
11. [11] KUANG L，TU S M，ZHANG Y Q，et al. Providing Privacy Preserving in Next POI Recommendation for Mobile Edge Computing [J/OL].Journal of Cloud Computing，2020，9（1）：10（2020-02-07）[2025-10-11].https://doi.org/10.1186/s13677-020-0158-3.
12. [12] XIE Q Q，ZHANG H，WANG L Q，et al. SecureLoc: A Fully Homomorphic Encryption-based Privacy Protection Scheme for Location-based Services [J/OL].Journal of Information Security and Applications，2025，93（C）：104190（2025-09-01）[2025-10-12].https://doi.org/10.1016/j.jisa.2025.104190.

About the authors

Peng Jing (1996-), female, Han ethnicity, from Anqing, Anhui, teaching assistant, master's degree, research direction: network information and security, artificial intelligence.