Graph Neural Network (GNN) and XAI for Adaptive, Cross-Institutional Fraud Detection

The rapid digitalization of financial services has significantly increased the scale and complexity of fraudulent activities across financial institutions. Conventional fraud detection systems, primarily based on rule-based mechanisms and tabular machine learning models, often fail to identify sophisticated and interconnected fraud schemes spanning multiple institutions. Furthermore, increasing regulatory requirements regarding transparency and data privacy restrict direct sharing of financial transaction data among organizations. This study proposes a privacy-preserving and adaptive fraud detection framework that integrates Graph Neural Networks (GNNs), Federated Learning (FL), and Explainable Artificial Intelligence (XAI) to enable collaborative fraud detection across institutions without compromising sensitive customer information.

The proposed framework models financial ecosystems as distributed heterogeneous graphs consisting of accounts, merchants, devices, and transactional relationships. A federated graph learning mechanism enables participating institutions to collaboratively train fraud detection models while preserving local data ownership. To address the interpretability limitations of deep learning systems in highly regulated financial environments, XAI techniques are integrated to provide transparent explanations for fraud predictions at transaction, account, and network levels. Additionally, adaptive learning strategies are incorporated to address evolving fraud patterns and concept drift over time.

The study presents the conceptual architecture, methodological framework, privacy-preserving mechanisms, explainability integration, and evaluation strategy for the proposed system. The framework aims to improve fraud detection accuracy, enhance regulatory compliance, and facilitate collaborative intelligence sharing among financial institutions while maintaining strict privacy guarantees.

Graph Neural Networks (GNN), Explainable Artificial Intelligence (XAI), Federated Learning, Financial Fraud Detection, Cross-Institutional Learning, Privacy-Preserving Machine Learning.

W. L. Hamilton, R. Ying, and J. Leskovec, “Inductive Representation Learning on Large Graphs,” in Advances in Neural Information Processing Systems (NeurIPS), Long Beach, CA, USA, 2017, pp. 1024–1034.

T. N. Kipf and M. Welling, “Semi-Supervised Classification with Graph Convolutional Networks,” in International Conference on Learning Representations (ICLR), Toulon, France, 2017.

P. Veličković, G. Cucurull, A. Casanova, A. Romero, P. Liò, and Y. Bengio, “Graph Attention Networks,” in International Conference on Learning Representations (ICLR), Vancouver, Canada, 2018.

R. Ying, D. Bourgeois, J. You, M. Zitnik, and J. Leskovec, “GNNExplainer: Generating Explanations for Graph Neural Networks,” in Advances in Neural Information Processing Systems (NeurIPS), vol. 32, Vancouver, Canada, 2019.

B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, “Communication-Efficient Learning of Deep Networks from Decentralized Data,” in Proc. 20th International Conference on Artificial Intelligence and Statistics (AISTATS), Fort Lauderdale, FL, USA, 2017, pp. 1273–1282.

Q. Yang, Y. Liu, T. Chen, and Y. Tong, “Federated Machine Learning: Concept and Applications,” ACM Transactions on Intelligent Systems and Technology, vol. 10, no. 2, pp. 1–19, Feb. 2019.

Z. Wu, S. Pan, F. Chen, G. Long, C. Zhang, and P. S. Yu, “A Comprehensive Survey on Graph Neural Networks,” IEEE Transactions on Neural Networks and Learning Systems, vol. 32, no. 1, pp. 4–24, Jan. 2021.

D. Zhou, J. Cui, H. Zhang, Z. Yang, C. Liu, and M. Sun, “Graph Neural Networks: A Review of Methods and Applications,” AI Open, vol. 1, pp. 57–81, 2020.

C. Agarwal, H. Lakkaraju, and M. Zitnik, “Evaluating Explainability for Graph Neural Networks,” Scientific Data, vol. 10, no. 1, pp. 1–21, 2023.

S. Kanamori, H. Watanabe, and T. Uchida, “Privacy-Preserving Federated Learning for Detecting Fraudulent Financial Transactions Across Multiple Banks,” IPSJ Transactions on Information and Systems, vol. 30, no. 4, pp. 789–800, 2022.

D. Cheng, Y. Zhang, and H. Liu, “Graph Neural Networks for Financial Fraud Detection: A Comprehensive Review,” ACM Computing Surveys, vol. 57, no. 4, pp. 1–35, 2024.

L. Ge, J. Han, and K. Xu, “A Review of Privacy-Preserving Research on Federated Graph Neural Networks,” Neurocomputing, vol. 575, Art. no. 128084, 2024.

S. Motie, J. Zhang, and X. Sun, “Financial Fraud Detection Using Graph Neural Networks,” Expert Systems with Applications, vol. 235, Art. no. 121118, 2024.

O. Owoade and K. Petersen, “Federated Graph Learning for Privacy-Preserving Financial Anomaly Detection,” Nature Communications, vol. 15, Art. no. 3456, 2024.

Y. Chen, X. Li, and T. Wang, “Deep Learning in Financial Fraud Detection: Innovations, Challenges, and Future Directions,” Discover Artificial Intelligence, vol. 5, no. 1, Art. no. 12, 2025.

S. K. Aljunaid, M. Al-Hadrani, and Y. H. Al-Mamary, “Explainable AI-Driven Federated Learning Model for Financial Fraud Detection,” Journal of Risk and Financial Management, vol. 18, no. 4, Art. no. 179, 2025.

M. Nandan, R. Gupta, and P. Singh, “A Survey of Graph Neural Networks for Explainable Artificial Intelligence,” Neural Computing and Applications, vol. 37, no. 9, pp. 14621–14655, 2025.

Z. Xia, J. Li, and R. Huang, “FinGraphFL: Financial Graph-Based Federated Learning with Differential Privacy for Credit Card Fraud Detection,” Mathematics, vol. 13, no. 9, Art. no. 1396, 2025.

Y. Liu, J. James, J. Kang, D. Niyato, and S. Zhang, “Privacy-Preserving Traffic Flow Prediction: A Federated Learning Approach,” IEEE Internet of Things Journal, vol. 7, no. 8, pp. 7751–7763, Aug. 2020.

J. Zhou, G. Cui, S. Hu, Z. Zhang, C. Yang, Z. Liu, L. Wang, C. Li, and M. Sun, “Graph Neural Networks: A Review of Methods and Applications,” AI Open, vol. 1, pp. 57–81, 2020.