AI-Driven Fraud Detection At Scale: A Novel Deep Learning Architecture For Securing High-Frequency Payment Networks
DOI:
https://doi.org/10.64882/ijrt.v13.i2.975Keywords:
fraud detection, temporal graph neural networks, attention mechanisms, payment security, anomaly detection, concept drift, scalable machine learningAbstract
The rapid expansion of instant payment systems and card-not-present transactions has introduced a pressing security imperative: the need for real-time fraud detection in high-throughput, low-latency payment networks. Traditional rule-based approaches and shallow machine learning models suffer from inherent shortcomings, including poor adaptability to concept drift, limited capacity to identify intricate fraud patterns, and challenges in scaling to handle vast transaction volumes (Chen et al., 2021; Rodriguez & Liu, 2023). These limitations stem primarily from their reliance on static, hand-crafted features and their failure to effectively capture complex temporal interdependencies in transaction sequences. In response to this gap, we introduce the Temporal Graph Attention Network with Anomaly-aware Embeddings (TGAT-AAE), a cutting-edge deep learning framework designed for scalable, adaptive fraud detection in financial ecosystems. TGAT-AAE incorporates three pivotal innovations: (1) a dynamic temporal graph representation that models evolving network topologies while maintaining sequential integrity; (2) an anomaly-aware embedding module employing contrastive learning to generate robust, discriminative features from highly imbalanced datasets; and (3) a multi-head temporal attention layer that focuses on anomalous sub-graphs, effectively filtering out benign patterns to enhance detection precision. Empirical assessments on the IEEE-CIS fraud detection benchmark and a proprietary sanitized dataset of over 50 million real-world transactions reveal that TGAT-AAE attains an F1-score of 0.87 and an AUC-PR of 0.92, outperforming leading baselines such as XGBoost, Isolation Forest, and GraphSAGE by 12-18% margins, all while maintaining inference latencies below 10 milliseconds (IEEE-CIS, 2019; Goyal & Ferrara, 2023). Furthermore, the model demonstrates exceptional resilience in concept drift simulations, where it sustains high accuracy amid shifting fraud tactics, and supports horizontal scaling to manage millions of transactions per second without performance degradation. By integrating self-supervised learning with graph-based temporal modeling, TGAT-AAE not only mitigates class imbalance but also enables continuous learning from streaming data, reducing the need for frequent retraining. This framework offers financial institutions a versatile, deployable solution to fortify payment infrastructures against adaptive adversarial threats, paving the way for more secure digital economies. Future work could explore integration with federated learning to enhance privacy-preserving capabilities across distributed banking networks.
References
Bank for International Settlements. (2022). Annual economic report 2022: The future of the monetary system. BIS Publications. https://www.bis.org/publ/arpdf/ar2022e.htm
Bhattacharyya, S., Jha, S., Tharakunnel, K., & Westland, J. C. (2011). Data mining for credit card fraud: A comparative study. Decision Support Systems, 50(3), 602–613. https://doi.org/10.1016/j.dss.2010.08.008 DOI: https://doi.org/10.1016/j.dss.2010.08.008
Bogen, M., Rieke, A., & Ahmed, S. (2022). Awareness in practice: Tensions in access to sensitive attribute data for antidiscrimination. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (pp. 1803–1815). Association for Computing Machinery. https://doi.org/10.1145/3531146.3533139 DOI: https://doi.org/10.1145/3531146.3533139
Bose, A., Bhattacharjee, K., & Roy, S. (2022). Adversarial attacks on graph neural networks for fraud detection: A comprehensive survey. ACM Computing Surveys, 55(4), Article 78, 1–38. https://doi.org/10.1145/3511808 DOI: https://doi.org/10.1145/3538704
Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16, 321–357. https://doi.org/10.1613/jair.953 DOI: https://doi.org/10.1613/jair.953
Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 785–794). Association for Computing Machinery. https://doi.org/10.1145/2939672.2939785 DOI: https://doi.org/10.1145/2939672.2939785
Chen, Y., Wang, X., & Zhang, L. (2021). Real-time fraud detection in payment networks: Challenges and opportunities. IEEE Transactions on Knowledge and Data Engineering, 33(8), 3127–3142. https://doi.org/10.1109/TKDE.2020.2975623
Dal Pozzolo, A., Caelen, O., Johnson, R. A., & Bontempi, G. (2015). Calibrating probability with undersampling for unbalanced classification. In Proceedings of the 2015 IEEE Symposium Series on Computational Intelligence (pp. 159–166). IEEE. https://doi.org/10.1109/SSCI.2015.33 DOI: https://doi.org/10.1109/SSCI.2015.33
Davis, J., & Goadrich, M. (2006). The relationship between precision-recall and ROC curves. In Proceedings of the 23rd International Conference on Machine Learning (pp. 233–240). Association for Computing Machinery. https://doi.org/10.1145/1143844.1143874 DOI: https://doi.org/10.1145/1143844.1143874
Deloitte. (2023). Global risk management survey: Financial services industry outlook. Deloitte Insights. https://www2.deloitte.com/global/en/insights/topics/risk-management.html
Fey, M., & Lenssen, J. E. (2019). Fast graph representation learning with PyTorch Geometric. In ICLR Workshop on Representation Learning on Graphs and Manifolds. https://arxiv.org/abs/1903.02428
FICO. (2023). FICO global fraud report: Payment fraud trends and countermeasures. FICO Analytics. https://www.fico.com/en/latest-thinking/report/global-fraud-report-2023
Fiore, U., De Santis, A., Perla, F., Zanetti, P., & Palmieri, F. (2019). Using generative adversarial networks for improving classification effectiveness in credit card fraud detection. Information Sciences, 479, 448–455. https://doi.org/10.1016/j.ins.2017.12.030 DOI: https://doi.org/10.1016/j.ins.2017.12.030
Goyal, P., & Ferrara, E. (2023). Graph neural networks for fraud detection: A comprehensive review. ACM Transactions on Intelligent Systems and Technology, 14(2), Article 32, 1–35. https://doi.org/10.1145/3570507 DOI: https://doi.org/10.1145/3623402
Hamilton, W. L., Ying, Z., & Leskovec, J. (2017). Inductive representation learning on large graphs. In Advances in Neural InformationProcessing Systems 30 (pp. 1024–1034). Curran Associates. https://proceedings.neurips.cc/paper/2017/file/5dd9db5e033da9c6fb5ba83c7a7ebea9-Paper.pdf
IEEE-CIS. (2019). IEEE-CIS fraud detection dataset [Data set]. Kaggle. https://www.kaggle.com/c/ieee-fraud-detection
Jurgovsky, J., Granitzer, M., Ziegler, K., Calabretto, S., Portier, P. E., He-Guelton, L., & Caelen, O. (2018). Sequence classification for credit-card fraud detection. Expert Systems with Applications, 100, 234–245. https://doi.org/10.1016/j.eswa.2018.01.037 DOI: https://doi.org/10.1016/j.eswa.2018.01.037
Kairouz, P., McMahan, H. B., Avent, B., Bellet, A., Bennis, M., Bhagoji, A. N., ... & Zhao, S. (2021). Advances and open problems in federated learning. Foundations and Trends in Machine Learning, 14(1–2), 1–210. https://doi.org/10.1561/2200000083 DOI: https://doi.org/10.1561/2200000083
Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., ... & Liu, T. Y. (2017). LightGBM: A highly efficient gradient boosting decision tree. In Advances in Neural Information Processing Systems 30 (pp. 3146–3154). Curran Associates. https://proceedings.neurips.cc/paper/2017/file/6449f44a102fde848669bdd9eb6b76fa-Paper.pdf
Kipf, T. N., & Welling, M. (2017). Semi-supervised classification with graph convolutional networks. In Proceedings of the 5th International Conference on Learning Representations. https://arxiv.org/abs/1609.02907
Kumar, S., Zhang, X., & Leskovec, J. (2022). Predicting dynamic embedding trajectory in temporal interaction networks. In Proceedings of the 28th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1269–1278). Association for Computing Machinery. https://doi.org/10.1145/3534678.3539102 DOI: https://doi.org/10.1145/3292500.3330895
Lin, T. Y., Goyal, P., Girshick, R., He, K., & Dollár, P. (2020). Focal loss for dense object detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(2), 318–327. https://doi.org/10.1109/TPAMI.2018.2858826 DOI: https://doi.org/10.1109/TPAMI.2018.2858826
Liu, Z., Chen, C., Yang, X., Zhou, J., Li, X., & Song, L. (2018). Heterogeneous graph neural networks for malicious account detection. In Proceedings of the 27th ACM International Conference on Information and Knowledge Management (pp. 2077–2085). Association for Computing Machinery. https://doi.org/10.1145/3269206.3272010 DOI: https://doi.org/10.1145/3269206.3272010
Loshchilov, I., & Hutter, F. (2019). Decoupled weight decay regularization. In Proceedings of the 7th International Conference on Learning Representations. https://arxiv.org/abs/1711.05101
Ma, Y., Guo, Z., Ren, Z., Tang, J., & Yin, D. (2020). Streaming graph neural networks. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 719–728). Association for Computing Machinery. https://doi.org/10.1145/3397271.3401092 DOI: https://doi.org/10.1145/3397271.3401092
McKinsey & Company. (2023). The 2023 McKinsey global payments report. McKinsey & Company Financial Services. https://www.mckinsey.com/industries/financial-services/our-insights/global-payments-report
McMahan, B., Moore, E., Ramage, D., Hampson, S., & Arcas, B. A. (2017). Communication-efficient learning of deep networks from decentralized data. In Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (pp. 1273–1282). PMLR. https://proceedings.mlr.press/v54/mcmahan17a.html
Nilson Report. (2024). Global card fraud losses. Nilson Report, Issue 1254. https://nilsonreport.com/mention/1254/1link/
Pang, G., Shen, C., Cao, L., & Hengel, A. V. D. (2021). Deep learning for anomaly detection: A review. ACM Computing Surveys, 54(2), Article 38, 1–38. https://doi.org/10.1145/3439950 DOI: https://doi.org/10.1145/3439950
Poursafaei, F., Huang, S., Pelrine, K., & Rabbany, R. (2022).Towards better evaluation for dynamic link prediction. In Advances in Neural Information Processing Systems 35 (pp. 32928–32941). Curran Associates. https://proceedings.neurips.cc/paper_files/paper/2022/file/d49042a5d49818711c401d34172f9900-Paper-Datasets_and_Benchmarks.pdf DOI: https://doi.org/10.52202/068431-2386
Pumsirirat, A., & Yan, L. (2020). Credit card fraud detection using deep learning based on auto-encoder and restricted Boltzmann machine. International Journal of Advanced Computer Science and Applications, 9(1), 18–25. https://doi.org/10.14569/IJACSA.2018.090103 DOI: https://doi.org/10.14569/IJACSA.2018.090103
Rodriguez, M., & Liu, H. (2023). Adaptive machine learning for evolving fraud patterns in digital payments. Journal of Financial Technology, 5(2), 112–134. https://doi.org/10.1016/j.jfintech.2023.100089
Rossi, E., Chamberlain, B., Frasca, F., Eynard, D., Monti, F., & Bronstein, M. (2020). Temporal graph networks for deep learning on dynamic graphs. In ICML 2020 Workshop on Graph Representation Learning. https://arxiv.org/abs/2006.10637
Roy, A., Sun, J., Mahoney, R., Alonzi, L., Adams, S., & Beling, P. (2018). Deep learning detecting fraud in credit card transactions. In Proceedings of the 2018 Systems and Information Engineering Design Symposium (pp. 129–134). IEEE. https://doi.org/10.1109/SIEDS.2018.8374722 DOI: https://doi.org/10.1109/SIEDS.2018.8374722
Saito, T., & Rehmsmeier, M. (2021). The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PLOS ONE, 10(3), Article e0118432. https://doi.org/10.1371/journal.pone.0118432 DOI: https://doi.org/10.1371/journal.pone.0118432
Shi, L., Wu, L., Ying, X., & Wu, X. (2022). Integrating threat intelligence for real-time fraud detection. IEEE Transactions on Dependable and Secure Computing, 19(4), 2534–2548. https://doi.org/10.1109/TDSC.2021.3089145
Statista. (2023). Digital payments worldwide: Statistics and market data. Statista Research Department. https://www.statista.com/outlook/dmo/fintech/digital-payments/worldwide
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., ... & Polosukhin, I. (2017). Attention is all you need. In Advances in Neural Information Processing Systems 30 (pp. 5998–6008). Curran Associates. https://proceedings.neurips.cc/paper/2017/file/3f5ee243547dee91fbd053c1c4a845aa-Paper.pdf
Veličković, P., Cucurull, G., Casanova, A., Romero, A., Liò, P., & Bengio, Y. (2018). Graph attention networks. In Proceedings of the 6th International Conference on Learning Representations. https://arxiv.org/abs/1710.10903
Wang, D., Lin, J., Cui, P., Jia, Q., Wang, Z., Fang, Y., ... & Zhu, W. (2019). A semi-supervised graph attentive network for financial fraud detection. In Proceedings of the 2019 IEEE International Conference on Data Mining (pp. 598–607). IEEE. https://doi.org/10.1109/ICDM.2019.00070 DOI: https://doi.org/10.1109/ICDM.2019.00070
Wang, M., Zheng, D., Ye, Z., Gan, Q., Li, M., Song, X., ... & Zhang, Z. (2020). Deep graph library: A graph-centric, highly-performant package for graph neural networks. arXiv preprint. https://arxiv.org/abs/1909.01315
Wang, X., Liu, Y., Zhang, C., & Yu, P. S. (2021). Heterogeneous graph attention network for fraud detection. In Proceedings of the Web Conference 2021 (pp. 1953–1964). Association for Computing Machinery. https://doi.org/10.1145/3442381.3449786 DOI: https://doi.org/10.1145/3442381.3449786
Weber, M., Domeniconi, G., Chen, J., Weidele, D. K. I., Bellei, C., Robinson, T., & Leiserson, C. E. (2019). Anti-money laundering in bitcoin: Experimenting with graph convolutional networks for financial forensics. In KDD 2019 Workshop on Anomaly Detection in Finance. https://arxiv.org/abs/1908.02591
Weinberger, K., Dasgupta, A., Langford, J., Smola, A., & Attenberg, J. (2009). Feature hashing for large scale multitask learning. In Proceedings of the 26th Annual International Conference on Machine Learning (pp. 1113–1120). Association for Computing Machinery. https://doi.org/10.1145/1553374.1553516 DOI: https://doi.org/10.1145/1553374.1553516
Whitrow, C., Hand, D. J., Juszczak, P., Weston, D., & Adams, N. M. (2020). Transaction aggregation as a strategy for credit card fraud detection. Data Mining and Knowledge Discovery, 18(1), 30–55. https://doi.org/10.1007/s10618-008-0116-z DOI: https://doi.org/10.1007/s10618-008-0116-z
Xu, D., Ruan, C., Korpeoglu, E., Kumar, S., & Achan, K. (2020). Inductive representation learning on temporal graphs. In Proceedings of the 8th International Conference on Learning Representations. https://arxiv.org/abs/2002.07962
Ying, R., Bourgeois, D., You, J., Zitnik, M., & Leskovec, J. (2019). GNNExplainer: Generating explanations for graph neural networks. In Advances in Neural Information Processing Systems 32 (pp. 9244–9255). Curran Associates. https://proceedings.neurips.cc/paper/2019/file/d80b7040b773199015de6d3b4293c8ff-Paper.pdf
Ying, R., He, R., Chen, K., Eksombatchai, P., Hamilton, W. L., & Leskovec, J. (2018). Graph convolutional neural networks for web-scale recommender systems. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 974–983). Association for Computing Machinery. https://doi.org/10.1145/3219819.3219890 DOI: https://doi.org/10.1145/3219819.3219890
Yoon, J., Jordon, J., & Van Der Schaar, M. (2018). GAIN: Missing data imputation using generative adversarial nets. In Proceedings of the 35th International Conference on Machine Learning (pp. 5689–5698). PMLR. https://proceedings.mlr.press/v80/yoon18a.html
Zeng, H., Zhou, H., Srivastava, A., Kanber, R., Prasad, M., & Hu, X. (2021). GraphSAINT: Graph sampling based inductive learning method. In Proceedings of the 9th International Conference on Learning Representations. https://arxiv.org/abs/1907.04931
Zhang, B. H., Lemoine, B., & Mitchell, M. (2018). Mitigating unwanted biases with adversarial learning. In *Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society* (pp. 335–340). Association for Computing Machinery. https://doi.org/10.1145/3278721.3278779 DOI: https://doi.org/10.1145/3278721.3278779
Zhang, Y., Liu, Q., & Wu, L. (2020). Neural network-based credit card fraud detection with feature engineering. IEEE Access, 8, 158275–158287. https://doi.org/10.1109/ACCESS.2020.3019867
Zheng, P., Yuan, S., Wu, X., Li, J., & Lu, A. (2022). Adversarial fraud detection: A survey of attacks and defenses. ACM Computing Surveys, 54(8), Article 167, 1–36. https://doi.org/10.1145/3466772 DOI: https://doi.org/10.1145/3466772
Downloads
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
