What question did this study set out to answer?

This research aims to enhance credit card fraud detection by leveraging graph-based techniques combined with machine learning methods.

May 16, 2026Open Access

Graph-Enhanced Fraud Detection: A Hybrid Graph Neural Network and XGBoost Framework for Credit Card Transaction Classification

Key Points

This research aims to enhance credit card fraud detection by leveraging graph-based techniques combined with machine learning methods.
Constructed a transaction graph with 10,000 nodes and 629,820 edges.
Employed two-layer mean-aggregation message passing for feature extraction.
Integrated graph features into XGBoost for classification.
Achieved an F1 score of 96.99%, compared to a baseline of 49.15%.
Demonstrated a +47.84% improvement in fraud detection performance.
Confirmed that graph features account for over 90% of predictive power.

Abstract

Credit card fraud detection remains one of the most pressing challenges in financial technology, costing institutions an estimated 32 billion annually. This paper proposes GNN-XGB, a hybrid framework combining Graph Neural Networks with XGBoost to exploit individual transaction features andgraph-neighborhood patterns. We construct a transaction graph of 10, 000 nodes and 629, 820 edges, apply two-layer mean-aggregation message passing, and feed graph features into XGBoost. Our framework achieves F1 Score of 96. 99% compared to 49. 15% baseline, a +47. 84% improvement. Graph featuresaccount for over 90% of predictive power, confirming fraud is a network phenomenon. Implementation uses pure Python with no specialized libraries.

Graph-Enhanced Fraud Detection: A Hybrid Graph Neural Network and XGBoost Framework for Credit Card Transaction Classification

Key Points

Abstract

Cite This Study