Privacy-Preserving Individual-Level COVID-19 Infection Prediction via Federated Graph Learning

TL;DR

This work tackles private, accurate infection prediction at the individual level by marrying federated learning with hypergraph neural networks. It introduces Falcon, which uses a spatio-temporal hypergraph and detached server-client propagation to model high-order user-location interactions without exposing raw trajectory data, and complements this with differential privacy and a plausible pseudo-location generator to guard against location inference. A macroscopic region-level model is coupled with the microscopic, mitigating utility loss caused by privacy tools. Across multi-scenario experiments, Falcon consistently outperforms baselines while offering quantifiable privacy protections, demonstrating practical impact for privacy-aware epidemic control and policy design.

Abstract

Accurately predicting individual-level infection state is of great value since its essential role in reducing the damage of the epidemic. However, there exists an inescapable risk of privacy leakage in the fine-grained user mobility trajectories required by individual-level infection prediction. In this paper, we focus on developing a framework of privacy-preserving individual-level infection prediction based on federated learning (FL) and graph neural networks (GNN). We propose Falcon, a Federated grAph Learning method for privacy-preserving individual-level infeCtion predictiON. It utilizes a novel hypergraph structure with spatio-temporal hyperedges to describe the complex interactions between individuals and locations in the contagion process. By organically combining the FL framework with hypergraph neural networks, the information propagation process of the graph machine learning is able to be divided into two stages distributed on the server and the clients, respectively, so as to effectively protect user privacy while transmitting high-level information. Furthermore, it elaborately designs a differential privacy perturbation mechanism as well as a plausible pseudo location generation approach to preserve user privacy in the graph structure. Besides, it introduces a cooperative coupling mechanism between the individual-level prediction model and an additional region-level model to mitigate the detrimental impacts caused by the injected obfuscation mechanisms. Extensive experimental results show that our methodology outperforms state-of-the-art algorithms and is able to protect user privacy against actual privacy attacks. Our code and datasets are available at the link: https://github.com/wjfu99/FL-epidemic.

Figures (9)

• Figure 1: Overall architecture of Falcon.
• Figure 2: Illustration of privacy-preserving FGML framework.
• Figure 3: Illustration of plausible pseudo location generation.
• Figure 4: Illustration of the coupling mechanism between microscopic and macroscopic models.
• Figure 5: The precision-recall (PR) trade-off curves of all baseline methods on the two scenarios.

Theorems & Definitions (2)

• Definition 2.1: Individual-level COVID-19 Infection Prediction in the manner of FL
• Definition 4.1: Disease Extinction Precision