Grad: Guided Relation Diffusion Generation for Graph Augmentation in Graph Fraud Detection

2025-12-19

AI Summary

In financial graph fraud detection, fraudsters evade detection via adaptive camouflage—mimicking benign user behavior—causing existing models to fail. Method: This paper proposes a relation-diffusion-based graph augmentation framework. It introduces a novel guided relation diffusion mechanism that jointly optimizes a learnable relation generator with supervised graph contrastive learning to explicitly model and amplify subtle behavioral discrepancies between fraudulent and legitimate users. Furthermore, pseudo-labeling is employed to construct homophilous edges, enhancing the propagation of fraud signals across the graph. Contribution/Results: Extensive experiments on the real-world WeChat Pay dataset and three public benchmarks demonstrate significant improvements over state-of-the-art methods: up to +11.10% in AUC and +43.95% in Average Precision.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Computer Vision: Diffusion Models for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Nowadays, Graph Fraud Detection (GFD) in financial scenarios has become an urgent research topic to protect online payment security. However, as organized crime groups are becoming more professional in real-world scenarios, fraudsters are employing more sophisticated camouflage strategies. Specifically, fraudsters disguise themselves by mimicking the behavioral data collected by platforms, ensuring that their key characteristics are consistent with those of benign users to a high degree, which we call Adaptive Camouflage. Consequently, this narrows the differences in behavioral traits between them and benign users within the platform's database, thereby making current GFD models lose efficiency. To address this problem, we propose a relation diffusion-based graph augmentation model Grad. In detail, Grad leverages a supervised graph contrastive learning module to enhance the fraud-benign difference and employs a guided relation diffusion generator to generate auxiliary homophilic relations from scratch. Based on these, weak fraudulent signals would be enhanced during the aggregation process, thus being obvious enough to be captured. Extensive experiments have been conducted on two real-world datasets provided by WeChat Pay, one of the largest online payment platforms with billions of users, and three public datasets. The results show that our proposed model Grad outperforms SOTA methods in both various scenarios, achieving at most 11.10% and 43.95% increases in AUC and AP, respectively. Our code is released at https://github.com/AI4Risk/antifraud and https://github.com/Muyiiiii/WWW25-Grad.

Problem

• Detects fraud in financial graphs with adaptive camouflage
• Enhances fraud-benign differences using guided relation diffusion
• Generates auxiliary homophilic relations to amplify weak signals

Innovation

• Supervised graph contrastive learning enhances fraud-benign differences
• Guided relation diffusion generator creates auxiliary homophilic relations
• Augmentation strengthens weak fraudulent signals for detection