Differentially Private Graph Diffusion with Applications in Personalized PageRanks

2024-06-22

arXiv.org

AI Summary

To address edge-level privacy leakage in graph data publishing—particularly in personalized PageRank (PPR) computation—this paper proposes the first differentially private graph diffusion framework. Methodologically, it extends Privacy Amplification by Iteration (PABI) to the Laplace noise mechanism for the first time; introduces an Infinity-Wasserstein distance analysis to substantially tighten privacy loss bounds across iterations; and designs a degree-aware thresholding mechanism to mitigate privacy-utility imbalance arising from high sensitivity of low-degree nodes. Experiments on real-world networks demonstrate that, under stringent privacy budgets (ε ≤ 1), the proposed method significantly outperforms existing baselines in PPR ranking quality, achieving both strong differential privacy guarantees and practical analytical utility.

Technology Category

• Machine Learning: Privacy
• Search and Optimization: Distributed Search
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Security and Privacy: Privacy-enhancing technologies
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems

Abstract

Graph diffusion, which iteratively propagates real-valued substances among the graph, is used in numerous graph/network-involved applications. However, releasing diffusion vectors may reveal sensitive linking information in the data such as transaction information in financial network data. However, protecting the privacy of graph data is challenging due to its interconnected nature. This work proposes a novel graph diffusion framework with edge-level differential privacy guarantees by using noisy diffusion iterates. The algorithm injects Laplace noise per diffusion iteration and adopts a degree-based thresholding function to mitigate the high sensitivity induced by low-degree nodes. Our privacy loss analysis is based on Privacy Amplification by Iteration (PABI), which to our best knowledge, is the first effort that analyzes PABI with Laplace noise and provides relevant applications. We also introduce a novel Infinity-Wasserstein distance tracking method, which tightens the analysis of privacy leakage and makes PABI more applicable in practice. We evaluate this framework by applying it to Personalized Pagerank computation for ranking tasks. Experiments on real-world network data demonstrate the superiority of our method under stringent privacy conditions.

Problem

• Graph Diffusion
• Privacy Protection
• Sensitive Information Leakage

Innovation

• Differential Privacy
• Graph Diffusion
• Privacy Leakage Monitoring