A Label-Free Heterophily-Guided Approach for Unsupervised Graph Fraud Detection

2025-02-18

AI Summary

This paper addresses the challenges of label scarcity and difficulty in modeling heterogeneous connection patterns between fraudsters and benign users in unsupervised graph-based fraud detection. To this end, we propose HUGE, a heterogeneity-guided framework. Its core contributions are: (1) HALO, the first label-free heterogeneity metric that quantifies structural dissimilarity among node neighborhoods without supervision; and (2) a joint MLP-GNN architecture incorporating ranking loss and asymmetric alignment loss to explicitly capture heterogeneous structural dependencies. Extensive experiments on six real-world graph datasets demonstrate that HUGE significantly outperforms existing unsupervised methods, validating the effectiveness of explicit heterogeneity modeling. The source code is publicly available.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Graph fraud detection (GFD) has rapidly advanced in protecting online services by identifying malicious fraudsters. Recent supervised GFD research highlights that heterophilic connections between fraudsters and users can greatly impact detection performance, since fraudsters tend to camouflage themselves by building more connections to benign users. Despite the promising performance of supervised GFD methods, the reliance on labels limits their applications to unsupervised scenarios; Additionally, accurately capturing complex and diverse heterophily patterns without labels poses a further challenge. To fill the gap, we propose a Heterophily-guided Unsupervised Graph fraud dEtection approach (HUGE) for unsupervised GFD, which contains two essential components: a heterophily estimation module and an alignment-based fraud detection module. In the heterophily estimation module, we design a novel label-free heterophily metric called HALO, which captures the critical graph properties for GFD, enabling its outstanding ability to estimate heterophily from node attributes. In the alignment-based fraud detection module, we develop a joint MLP-GNN architecture with ranking loss and asymmetric alignment loss. The ranking loss aligns the predicted fraud score with the relative order of HALO, providing an extra robustness guarantee by comparing heterophily among non-adjacent nodes. Moreover, the asymmetric alignment loss effectively utilizes structural information while alleviating the feature-smooth effects of GNNs.Extensive experiments on 6 datasets demonstrate that HUGE significantly outperforms competitors, showcasing its effectiveness and robustness. The source code of HUGE is at https://github.com/CampanulaBells/HUGE-GAD.

Problem

• Detects fraud without labeled data using graph heterophily
• Estimates heterophily with a novel label-free metric HALO
• Aligns fraud scores with heterophily for robust detection

Innovation

• Label-free heterophily metric HALO
• Joint MLP-GNN architecture
• Asymmetric alignment loss utilization