融合图神经网络和深度图聚类的联邦推荐算法

doi:10.3969/j.issn.0255-8297.2026.01.006

摘要/Abstract

摘要： 联邦学习作为解决推荐系统隐私安全问题的主流框架，在实际应用中却面临推荐精度欠佳、隐私保护力度不足及通信开销过大的问题。针对这些问题，本文提出一种融合图神经网络与深度图聚类的联邦推荐算法。首先，利用图神经网络对复杂的用户-项目的高阶交互关系进行捕捉，以提升推荐系统的推荐精度；其次，在联邦学习客户端与服务器端的通信环节注入差分隐私噪声以模糊真实梯度，进而增强推荐系统的隐私保护能力；最后，通过引入深度图聚类对客户端实施聚类，选取各簇的客户端代表参与训练，并将所得参数在簇内共享，以加快模型收敛速度，降低联邦学习框架下的通信开销。基于真实数据集的实验结果表明，所提算法在提高推荐精度的同时，能够增强系统的隐私保护力度并减少通信开销。

关键词: 推荐系统, 联邦学习, 隐私保护, 深度图聚类, 图神经网络

Abstract: Federated learning, as the main framework for addressing privacy and security issues in recommendation systems, faces problems such as poor recommendation accuracy, insufficient privacy protection, and excessive communication overhead in practical applications. To address these issues, this paper proposed a federated recommendation algorithm integrating graph neural networks and depth graph clustering. Firstly, a graph neural network was used to capture high-order complex user and item interaction relationships, improving the recommendation accuracy of the recommendation system. Secondly, differential privacy noise was injected into the communication link between the federated learning clients and server to blur the true gradient, thereby enhancing the privacy protection capability of the recommendation system. Finally, clients were clustered by introducing depth graph clustering, and client representatives from each cluster were selected to participate in training. The obtained parameters were shared within the cluster to accelerate the convergence speed of the model and reduce communication overhead under the federated learning framework. The experimental results on real datasets show that the proposed algorithm can enhance privacy protection of the system and reduce communication overhead while improving recommendation accuracy.

Key words: recommendation system, federated learning, privacy protection, depth graph clustering, graph neural network

中图分类号:

TP391

伊华伟, 宋仕玺, 王艳飞, 白思怡. 融合图神经网络和深度图聚类的联邦推荐算法[J]. 应用科学学报, 2026, 44(1): 83-96.

YI Huawei, SONG Shixi, WANG Yanfei, BAI Siyi. Federated Recommendation Algorithm Integrating Graph Neural Networks and Depth Graph Clustering[J]. Journal of Applied Sciences, 2026, 44(1): 83-96.

参考文献

[1] 代宝, 续杨晓雪, 罗蕊. 社交媒体用户信息过载的影响因素及其后果[J]. 现代情报, 2020, 40(1): 152-158. Dai B, Xu Y X X, Luo R. Review on the antecedents and consequences of social media users’ information overload [J]. Journal of Modern Information, 2020, 40(1): 152-158. (in Chinese)
[2] 郭桃林. 面向协同过滤推荐的差分隐私保护技术研究[D]. 南京: 东南大学, 2020.
[3] Wu S W, Sun F, Zhang W T, et al. Graph neural networks in recommender systems: a survey [J]. ACM Computing Surveys, 2023, 55(5): 1-37.
[4] 赵若宇. 图神经网络在推荐系统中的应用[D]. 合肥: 中国科学技术大学, 2021.
[5] Muhammad K, Wang Q Q, O’Reilly-Morgan D, et al. FedFast: going beyond average for faster training of federated recommender systems [C]//26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2020: 1234-1242.
[6] Wang X, He X N, Wang M, et al. Neural graph collaborative filtering [C]//42nd International ACM SIGIR Conference on Research and Development in Information Retrieval, 2019: 165-174.
[7] Liu Z, Meng L, Zhang J, et al. Deoscillated graph collaborative filtering [C]//Topological, Algebraic and Geometric Learning Workshops, 2022: 248-257.
[8] Fan W Q, Ma Y, Li Q, et al. A graph neural network framework for social recommendations [J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 34(5): 2033-2047.
[9] Dawn S, Das M, Bandyopadhyay S. SoURA: a user-reliability-aware social recommendation system based on graph neural network [J]. Neural Computing and Applications, 2023, 35(25): 18533-18551.
[10] Gao Q, Ma P C. Graph neural network and context-aware based user behavior prediction and recommendation system research [J]. Computational Intelligence and Neuroscience, 2020(1): 8812370.
[11] Zhao Y F, Ou M D, Zhang R Z, et al. Attributed graph neural networks for recommendation systems on large-scale and sparse graph [DB/OL]. (2024-12-26) [2025-08-12]. https://arxiv.org/abs/2112.1389.
[12] Mcmahan H B, Moore E, Ramage D, et al. Communication-efficient learning of deep networks from decentralized data [C]//International Conference on Artificial Intelligence and Statistics, 2016:1-11.
[13] Ammad-ud-din M, Ivannikova E, Khan S A, et al. Federated collaborative filtering for privacy-preserving personalized recommendation system [DB/OL]. (2019-01-29) [2025-08-12]. https://arxiv.org/abs/1901.09888.
[14] Lin G Y, Liang F, Pan W K, et al. FedRec: federated recommendation with explicit feedback [J]. IEEE Intelligent Systems, 2021, 36(5): 21-30.
[15] Chai D, Wang L Y, Chen K, et al. Secure federated matrix factorization [J]. IEEE Intelligent Systems, 2021, 36(5): 11-20.
[16] Wu C H, Wu F Z, Lyu L J, et al. A federated graph neural network framework for privacypreserving personalization [J]. Nature Communications, 2022, 13: 3091.
[17] Liu Z W, Yang L W, Fan Z W, et al. Federated social recommendation with graph neural network [J]. ACM Transactions on Intelligent Systems and Technology, 2022, 13(4): 1-24.
[18] Chen Y C, Feng C Y, Feng D Q. Privacy-preserving hierarchical federated recommendation systems [J]. IEEE Communications Letters, 2023, 27(5): 1312-1316.
[19] Ali W, Kumar R, Deng Z Y, et al. A federated learning approach for privacy protection in context-aware recommender systems [J]. The Computer Journal, 2021, 64(7): 1016-1027.
[20] Li Z, Bilal M, Xu X L, et al. Federated learning-based cross-enterprise recommendation with graph neural networks [J]. IEEE Transactions on Industrial Informatics, 2023, 19(1): 673-682.
[21] Perifanis V, Drosatos G, Stamatelatos G, et al. FedPOIRec: privacy-preserving federated POI recommendation with social influence [J]. Information Sciences, 2023, 623: 767-790.
[22] Ribero M, Henderson J, Williamson S, et al. Federating recommendations using differentially private prototypes [J]. Pattern Recognition, 2022, 129: 108746.