What question did this study set out to answer?

To develop an efficient and interpretable system for real-time financial fraud detection using advanced machine learning techniques.

March 3, 2026Open Access

Real-Time Transaction Fraud Detection Using Ensemble Learning and Graph Neural Networks

Key Points

To develop an efficient and interpretable system for real-time financial fraud detection using advanced machine learning techniques.
Integrated Isolation Forest, XGBoost, and Graph Neural Networks in a weighted ensemble framework.
Implemented SHAP-based explanations for interpreting model outputs.
Introduced a concept drift monitoring mechanism to adapt to evolving fraud strategies.
Conducted experimental evaluations to assess performance metrics.
Achieved a ROC-AUC score of 0.94 indicating high detection accuracy.
Maintained inference latency of less than 250 milliseconds for real-time operation.
Demonstrated capability to identify both behavioral anomalies and relational fraud patterns.

Abstract

Financial fraud detection requires scalable and interpretable solutions capable of operating in real time. This work presents a production-oriented fraud detection framework integrating Isolation Forest, XGBoost, and Graph Neural Networks within a weighted ensemble architecture. The proposed system detects behavioral anomalies and relational fraud patterns while maintaining interpretability using SHAP-based explanations. A concept drift monitoring mechanism ensures long-term robustness against evolving fraud strategies. Experimental evaluation demonstrates a ROC-AUC of 0.94 with sub-250ms inference latency, highlighting the framework’s suitability for real-world financial deployment

Real-Time Transaction Fraud Detection Using Ensemble Learning and Graph Neural Networks

Key Points

Abstract

Cite This Study