Federated Graph Analytics with Differential Privacy

TL;DR

This paper introduces FEAT and FEAT+ frameworks for federated graph analytics that preserve differential privacy, enabling collaborative analysis with significantly improved utility over baseline methods.

Contribution

The paper proposes the first federated graph analytics framework, FEAT, and an optimized version, FEAT+, which effectively balance privacy and utility in collaborative graph analysis.

Findings

FEAT and FEAT+ outperform baseline approaches by up to four orders of magnitude.

FEAT+ leverages local subgraphs for improved utility.

Extensive experiments validate the effectiveness of the proposed frameworks.

Abstract

Collaborative graph analysis across multiple institutions is becoming increasingly popular. Realistic examples include social network analysis across various social platforms, financial transaction analysis across multiple banks, and analyzing the transmission of infectious diseases across multiple hospitals. We define the federated graph analytics, a new problem for collaborative graph analytics under differential privacy. Although differentially private graph analysis has been widely studied, it fails to achieve a good tradeoff between utility and privacy in federated scenarios, due to the limited view of local clients and overlapping information across multiple subgraphs. Motivated by this, we first propose a federated graph analytic framework, named FEAT, which enables arbitrary downstream common graph statistics while preserving individual privacy. Furthermore, we introduce an optimized framework based on our proposed degree-based partition algorithm, called FEAT+, which improves the overall utility by leveraging the true local subgraphs. Finally, extensive experiments demonstrate that our FEAT and FEAT+ significantly outperform the baseline approach by approximately one and four orders of magnitude, respectively.