《计算机应用》唯一官方网站 ›› 2025, Vol. 45 ›› Issue (1): 136-143.DOI: 10.11772/j.issn.1001-9081.2024010044

• 网络空间安全 • 上一篇    下一篇

基于联邦图神经网络的位置隐私保护推荐方案

朱亮1, 慕京哲1, 左洪强2, 谷晶中2(), 朱付保1   

  1. 1.郑州轻工业大学 计算机科学与技术学院,郑州 450001
    2.山谷网安科技股份有限公司,郑州 450000
  • 收稿日期:2024-01-17 修回日期:2024-03-27 接受日期:2024-03-27 发布日期:2024-05-09 出版日期:2025-01-10
  • 通讯作者: 谷晶中
  • 作者简介:朱亮(1987—),男,河南焦作人,副教授,博士,主要研究方向:智能推荐、隐私保护;
    慕京哲(1999—),男,河南焦作人,硕士研究生,主要研究方向:智能推荐、隐私保护;
    左洪强(1989—),男,河南郑州人,主要研究方向:网络安全;
    朱付保(1974—),男,河南郑州人,教授,博士,主要研究方向:地理信息系统、人工智能。
  • 基金资助:
    国家自然科学基金资助项目(61902361);河南省重点研发专项(221111210500);河南省研究生联合培养基地项目(YJS2022JD08)

Location privacy-preserving recommendation scheme based on federated graph neural network

Liang ZHU1, Jingzhe MU1, Hongqiang ZUO2, Jingzhong GU2(), Fubao ZHU1   

  1. 1.School of Computer Science and Technology,Zhengzhou University of Light Industry,Zhengzhou Henan 450001,China
    2.Shangu Cyber Security Technology Company Limited,Zhengzhou Henan 450000,China
  • Received:2024-01-17 Revised:2024-03-27 Accepted:2024-03-27 Online:2024-05-09 Published:2025-01-10
  • Contact: Jingzhong GU
  • About author:ZHU Liang, born in 1987, Ph. D., associate professor. His research interests include smart recommendation, privacy protection.
    MU Jingzhe, born in 1999, M. S. candidate. His research interests include smart recommendation, privacy protection.
    ZUO Hongqiang, born in 1989. His research interests include network security.
    ZHU Fubao, born in 1974, Ph. D., professor. His research interests include geographic information system, artificial intelligence.
  • Supported by:
    National Natural Science Foundation of China(61902361);Henan Province Key Research and Development Special Project(221111210500);Henan Postgraduate Joint Training Base Project(YJS2022JD08)

摘要:

传统的位置服务推荐方案未考虑用户偏好及潜在社交关系等问题,导致推荐结果无法满足用户的个性化需求。图神经网络(GNN)凭借较好的图结构数据处理能力,被广泛用于位置推荐领域;然而,此前研究里中心化的数据范式容易造成位置隐私泄露的问题。因此,提出一种基于联邦图神经网络的位置隐私保护推荐方案(FedGNN-LPR)。首先,通过图注意力网络学习用户的社交关系嵌入和兴趣点(POI)嵌入;其次,建立基于POI的伪标签标注模型预测用户对未知位置的访问次数,以保护用户隐私并缓解冷启动问题;最后,提出基于差分隐私的聚类联邦学习策略保护客户端的交互数据并解决数据异质性问题。在两个公开的真实数据集上进行实验的结果表明,在平均绝对值误差(MAE)和均方根误差(RMSE)方面,所提方案比联邦平均(FedAvg)算法分别降低了7.89%和9.29%,比FL+HC算法分别降低了2.32%和2.75%;并且,所提方案在联邦学习位置推荐上展现出更好的性能。因此,FedGNN-LPR不仅能保护用户位置隐私,而且提高了位置推荐性能。

关键词: 基于位置的社交网络, 联邦学习, 图注意力网络, 伪标签, 位置推荐

Abstract:

Traditional location service recommendation schemes lack consideration of user preferences and potential social relationships, resulting in recommendation results that fail to meet user’s personalized needs. Graph Neural Networks (GNNs) are widely used in the field of location recommendation by its good graph structure data processing capabilities. However, the previous studies’ centralized data paradigm is easy to lead to the issue of location privacy leakage. Therefore, a Location Privacy-preserving Recommendation scheme based on Federated Graph Neural Network (FedGNN-LPR) was proposed. Firstly, the user’s social relationship embedding and Point-Of-Interest (POI) embedding were learned through the graph attention network. Secondly, a POI-based pseudo labelling model was developed to predict the number of user visits to an unknown location, so as to protect user privacy and alleviate the cold-start problem. Finally, a clustered federated learning strategy based on differential privacy was proposed to protect client interaction data and solve the problem of data heterogeneity. Experiments were conducted on two publicly available real datasets, and the results demonstrate that the proposed scheme is reduced by 7.89% and 9.29% respectively compared to the Federated Averaging (FedAvg) algorithm, and 2.32% and 2.75% respectively compared to the FL+HC algorithm, in terms of Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). Moreover, it is shown that FedGNN-LPR exhibits better performance on federated learning location recommendation. Therefore, FedGNN-LPR not only protects user location privacy, but also improves location recommendation performance.

Key words: Location-Based Social Network (LBSN), federated learning, graph attention network, pseudo label, location recommendation

中图分类号: