Detecting Credit Card Fraud via Heterogeneous Graph Neural Networks with Graph Attention

2025-04-11

AI Summary

Addressing the challenges of sparse fraudulent samples, complex transactional relationships, and temporal sensitivity in credit card fraud detection, this paper constructs a user-merchant-transaction heterogeneous graph. It proposes the first integration of heterogeneous graph neural networks (HGNNs) with dynamic graph attention mechanisms and time-decay modeling to explicitly capture high-order structural dependencies and temporal evolution patterns. To mitigate extreme class imbalance, SMOTE-based oversampling and a cost-sensitive loss function are further incorporated. Experiments on the IEEE-CIS dataset demonstrate that the method significantly outperforms baseline GNNs—including GCN, GAT, and GraphSAGE—achieving consistent improvements in both accuracy and OC-ROC score. These results validate the effectiveness and advancement of heterogeneous graph representation learning combined with multi-strategy optimization for fraud detection.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

This study proposes a credit card fraud detection method based on Heterogeneous Graph Neural Network (HGNN) to address fraud in complex transaction networks. Unlike traditional machine learning methods that rely solely on numerical features of transaction records, this approach constructs heterogeneous transaction graphs. These graphs incorporate multiple node types, including users, merchants, and transactions. By leveraging graph neural networks, the model captures higher-order transaction relationships. A Graph Attention Mechanism is employed to dynamically assign weights to different transaction relationships. Additionally, a Temporal Decay Mechanism is integrated to enhance the model's sensitivity to time-related fraud patterns. To address the scarcity of fraudulent transaction samples, this study applies SMOTE oversampling and Cost-sensitive Learning. These techniques strengthen the model's ability to identify fraudulent transactions. Experimental results demonstrate that the proposed method outperforms existing GNN models, including GCN, GAT, and GraphSAGE, on the IEEE-CIS Fraud Detection dataset. The model achieves notable improvements in both accuracy and OC-ROC. Future research may explore the integration of dynamic graph neural networks and reinforcement learning. Such advancements could enhance the real-time adaptability of fraud detection systems and provide more intelligent solutions for financial risk control.

Problem

• Detect credit card fraud using heterogeneous graph neural networks
• Address fraud in complex transaction networks with multiple node types
• Improve accuracy and OC-ROC over traditional GNN models

Innovation

• HGNN for complex transaction fraud detection
• Graph Attention Mechanism for dynamic weighting
• SMOTE and Cost-sensitive Learning for imbalance