Graph Neural Networks for Fraudulent Transaction Pattern Detection in Digital Lending Systems

Kaleshwar Aryasomayajula

Kaleshwar Aryasomayajula Sr Specialist Software Developer, Charles Schwab, San Tan Valley, Arizona, USA

Keywords: graph neural networks, digital lending, fraud detection, transaction patterns, credit risk, graph anomaly detection, online lending, imbalanced data, underbanked borrowers, decision governance

Abstract

Digital lending systems handle borrower applications, repayment events, device traces, and payment flows in environments where fraud arises from relationships among actors. A single application record rarely exposes coordinated fraud, while shared devices, repeated accounts, recycled contacts, and post-disbursement transfers often reveal hidden links. This article examines the use of Graph Neural Networks for detecting fraudulent transaction patterns in digital lending, with particular attention to underbanked and thin-file borrowers. The study aims to build an analytical model that connects graph construction, fraud pattern recognition, and decision governance. The material base covers recent studies on financial fraud detection, consumer loan fraud, graph anomaly detection, imbalanced learning and online credit loan risk. Comparative source analysis, conceptual synthesis, and typologization guide the research design. The article identifies graph representation principles, GNN mechanisms for coordinated fraud detection, and implementation rules for credit decision workflows. The proposed model supports auditable fraud controls without replacing creditworthiness assessment.

References

1. Cherif, A., Ammar, H., Kalkatawi, M., Alshehri, S., & Imine, A. (2024). Encoder-decoder graph neural network for credit card fraud detection. Journal of King Saud University - Computer and Information Sciences, 36(3), Article 102003. https://doi.org/10.1016/j.jksuci.2024.102003
2. Cheng, D., Zou, Y., Xiang, S., & Jiang, C. (2025). Graph neural networks for financial fraud detection: A review. Frontiers of Computer Science. https://doi.org/10.1007/s11704-024-40474-y
3. Huang, M., Liu, Y., Ao, X., Li, K., Chi, J., Feng, J., Yang, H., & He, Q. (2022). AUC-oriented graph neural network for fraud detection. In Proceedings of the ACM Web Conference 2022 (pp. 1311-1321). Association for Computing Machinery. https://doi.org/10.1145/3485447.3512178
4. Huang, X., Yang, Y., Wang, Y., Wang, C., Zhang, Z., Xu, J., Chen, L., & Vazirgiannis, M. (2022). DGraph: A large-scale financial dataset for graph anomaly detection. Advances in Neural Information Processing Systems, 35, 22765-22777.
5. Lou, C., Wang, Y., Li, J., Qian, Y., & Li, X. (2025). Graph neural network for fraud detection via context encoding and adaptive aggregation. Expert Systems with Applications, 261, Article 125473. https://doi.org/10.1016/j.eswa.2024.125473
6. Motie, S., & Raahemi, B. (2024). Financial fraud detection using graph neural networks: A systematic review. Expert Systems with Applications, 240, Article 122156. https://doi.org/10.1016/j.eswa.2023.122156
7. Tang, H., Wang, C., Zheng, J., & Jiang, C. (2024). Enabling graph neural networks for semi-supervised risk prediction in online credit loan services. ACM Transactions on Intelligent Systems and Technology, 15(1), Article 13, 1-24. https://doi.org/10.1145/3623401
8. Tong, G., & Shen, J. (2023). Financial transaction fraud detector based on imbalance learning and graph neural network. Applied Soft Computing, 149, Article 110984. https://doi.org/10.1016/j.asoc.2023.110984
9. Wen, J., Tang, X., & Lu, J. (2024). An imbalanced learning method based on graph tran-smote for fraud detection. Scientific Reports, 14, Article 16560. https://doi.org/10.1038/s41598-024-67550-4
10. Xu, B., Shen, H., Sun, B., An, R., Cao, Q., & Cheng, X. (2021). Towards consumer loan fraud detection: Graph neural networks with role-constrained conditional random field. Proceedings of the AAAI Conference on Artificial Intelligence, 35(5), 4537-4545. https://doi.org/10.1609/aaai.v35i5.16582