HoGS: Homophily-Oriented Graph Synthesis for Local Differentially Private GNN Training

2026-02-09

AI Summary

This work addresses the significant performance degradation of graph neural networks (GNNs) under local differential privacy (LDP) by proposing the HoGS framework. HoGS leverages graph homophily to jointly reconstruct both graph structure and node features from LDP-perturbed data, generating high-quality synthetic graphs for GNN training while rigorously preserving the privacy of node attributes and links. To the best of our knowledge, this is the first approach to integrate graph homophily modeling with LDP mechanisms, enabling unified protection of these two types of sensitive information and substantially reducing utility loss. Experimental results on three real-world datasets demonstrate that various state-of-the-art GNN models trained on synthetic graphs produced by HoGS achieve significantly higher accuracy than those trained using existing LDP baselines.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Humans and AI: Human-in-the-loop Machine Learning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Security and Privacy: Data transparency and provenance
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Graph neural networks (GNNs) have demonstrated remarkable performance in various graph-based machine learning tasks by effectively modeling high-order interactions between nodes. However, training GNNs without protection may leak sensitive personal information in graph data, including links and node features. Local differential privacy (LDP) is an advanced technique for protecting data privacy in decentralized networks. Unfortunately, existing local differentially private GNNs either only preserve link privacy or suffer significant utility loss in the process of preserving link and node feature privacy. In this paper, we propose an effective LDP framework, called HoGS, which trains GNNs with link and feature protection by generating a synthetic graph. Concretely, HoGS first collects the link and feature information of the graph under LDP, and then utilizes the phenomenon of homophily in graph data to reconstruct the graph structure and node features separately, thereby effectively mitigating the negative impact of LDP on the downstream GNN training. We theoretically analyze the privacy guarantee of HoGS and conduct experiments using the generated synthetic graph as input to various state-of-the-art GNN architectures. Experimental results on three real-world datasets show that HoGS significantly outperforms baseline methods in the accuracy of training GNNs.

Problem

• Graph Neural Networks
• Local Differential Privacy
• Privacy Preservation
• Graph Data
• Utility Loss

Innovation

• Local Differential Privacy
• Graph Neural Networks
• Homophily
• Graph Synthesis
• Privacy-Preserving Machine Learning