Graph Neural Network-Based Framework for Real-Time Financial Fraud Detection in Digital Payment Ecosystems
DOI:
https://doi.org/10.71426/jcdt.v1.i2.pp91-97Keywords:
Graph Neural Networks, Financial fraud detection, Digital payment ecosystems, Graph Attention NetworksAbstract
The rapid growth of digital payment ecosystems, which include mobile wallets, UPI platforms, online banking channels, and merchant-integrated gateways, has led to a big rise in both the number of transactions and the complexity of financial fraud. Standard systems for finding fraud mostly use rule-based methods or independent transaction-level classifiers. These systems can't pick up on the complicated relational and temporal dependencies that are common in modern fraud patterns. This research presents a Graph Neural Network (GNN)-based real-time fraud detection framework that conceptualizes the digital payment ecosystem as a dynamic, multi-relational graph consisting of users, devices, merchants, IP addresses, and transactional interactions. The framework combines Relational Graph Convolutional Networks, Graph Attention Networks, and Temporal Graph Networks to learn behavioral patterns that change over time, in context, and in structure. A hybrid supervised–unsupervised scoring module is used to figure out how likely fraud is. This module can find both known and new types of attacks. A lot of tests on big synthetic payment datasets show that the proposed model does much better than traditional machine learning and deep learning baselines in terms of AUC-ROC, precision, and F1-score. It also has an inference latency of less than 50 ms, which makes it good for real-time use. The results show that GNN-based methods are effective for next-generation financial systems that need to be safe, scalable, and smart when it comes to stopping fraud.
References
[1] Zhao S. Design of machine learning-based credit card fraud detection model. In: Proceedings of the 2025 IEEE 3rd International Conference on Sensors, Electronics and Computer Engineering (ICSECE); 29 Aug 2025; pp. 1327–1330. Available from: https://ieeexplore.ieee.org/document/11256940
[2] Carcillo F, Dal Pozzolo A, Le Borgne YA, Caelen O, Mazzer Y, Bontempi G. Scarff: A scalable framework for streaming credit card fraud detection with Spark. Information Fusion. 2018 May;41:182–194. Available from: https://doi.org/10.1016/j.inffus.2017.09.005
[3] Whitrow C, Hand DJ, Juszczak P, Weston D, Adams NM. Transaction aggregation as a strategy for credit card fraud detection. Data Mining and Knowledge Discovery. 2009;18(1):30–55. Available from: https://doi.org/10.1007/s10618-008-0116-z
[4] Sahin Y, Bulkan S, Duman E. A cost-sensitive decision tree approach for fraud detection. Expert Systems with Applications. 2013 Nov;40(15):5916–5923. Available from: https://doi.org/10.1016/j.eswa.2013.05.021
[5] Cao L, Ou Y, Yu PS. Coupled behavior analysis with applications. IEEE Transactions on Knowledge and Data Engineering. 2012 Aug;24(8):1378–1392. Available from: https://doi.org/10.1109/TKDE.2011.129
[6] Akoglu L, Tong H, Koutra D. Graph-based anomaly detection and description: A survey. Data Mining and Knowledge Discovery. 2015 May;29(3):626–688. Available from: https://doi.org/10.1007/s10618-014-0365-y
[7] Kipf TN, Welling M. Semi-supervised classification with graph convolutional networks. In: Proceedings of the International Conference on Learning Representations (ICLR); 2017. arXiv:1609.02907. Available from: https://arxiv.org/abs/1609.02907
[8] Hamilton W, Ying Z, Leskovec J. Inductive representation learning on large graphs. In: Advances in Neural Information Processing Systems (NeurIPS); 2017; Vol. 30. Available from: https://proceedings.neurips.cc/paper/2017/hash/5dd9db5e033da9c6fb5ba83c7a7ebea9-Abstract.html
[9] Cao B, Mao M, Viidu S. Fraud detection in heterogeneous information networks. United States Patent US10460320B1; 29 Oct 2019. Available from: https://patents.google.com/patent/US10460320B1
[10] Schlichtkrull M, Kipf TN, Bloem P, van den Berg R, Titov I, Welling M. Modeling relational data with graph convolutional networks. In: Proceedings of the European Semantic Web Conference (ESWC); 3 Jun 2018; pp. 593–607. Available from: https://doi.org/10.1007/978-3-319-93417-4_38
[11] Veličković P, Cucurull G, Casanova A, Romero A, Liò P, Bengio Y. Graph attention networks. arXiv preprint. 2017 Oct; arXiv:1710.10903. Available from: https://arxiv.org/abs/1710.10903
[12] Rossi E, Chamberlain B, Frasca F, Eynard D, Monti F, Bronstein M. Temporal graph networks for deep learning on dynamic graphs. arXiv preprint. 2020 Jun; arXiv:2006.10637. Available from: https://arxiv.org/abs/2006.10637
[13] Wang H, Abraham Z. Concept drift detection for streaming data. In: Proceedings of the International Joint Conference on Neural Networks (IJCNN); 12 Jul 2015; pp. 1–9. Available from: https://doi.org/10.1109/IJCNN.2015.7280398
[14] Brommer C, Jung R, Steinbrener J, Weiss S. MaRS: A modular and robust sensor-fusion framework. IEEE Robotics and Automation Letters. 2021 Apr;6(2):359–366. Available from: https://doi.org/10.1109/LRA.2020.3043195
[15] Juszczak P, Adams NM, Hand DJ, Whitrow C, Weston D. Off-the-peg and bespoke classifiers for fraud detection. Computational Statistics & Data Analysis. 2008 Aug;52(9):4521–4532. Available from: https://doi.org/10.1016/j.csda.2008.03.021
[16] Dou Y, Liu Z, Sun L, Deng Y, Peng H, Yu PS. Enhancing graph neural network-based fraud detectors against camouflaged fraudsters. In: Proceedings of the 29th ACM International Conference on Information and Knowledge Management (CIKM); 2020. Available from: https://doi.org/10.1145/3340531.3411903
[17] Lu M, Han Z, Rao SX, Zhang Z, Zhao Y, Shan Y, et al. BRIGHT: Graph neural networks in real-time fraud detection. In: Proceedings of the 31st ACM International Conference on Information and Knowledge Management (CIKM); 16 Oct 2022; pp. 3342–3351. Available from: https://doi.org/10.1145/3511808.3557136
[18] Xu D, Ruan Y, Korpeoglu E, Kumar S, Achan K. Inductive representation learning on temporal graphs. In: Proceedings of the International Conference on Learning Representations (ICLR); 2020. Available from: https://arxiv.org/abs/2002.07962
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Ramakrishna Penaganti (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
