Adversarially Robust and Explainable AI for Real-Time Financial Fraud Detection in High-Frequency Transactions
Main Article Content
Abstract
The rapid expansion of digital financial transactions has led to an increase in sophisticated fraud techniques, particularly within high-frequency trading and online banking systems. Many existing fraud detection models struggle to detect adversarially crafted fraudulent transactions, where malicious actors exploit system vulnerabilities to evade detection. Additionally, traditional artificial intelligence (AI)-based fraud detection mechanisms often lack transparency, making it difficult for financial analysts and regulatory bodies to understand how decisions are made. This lack of interpretability raises concerns regarding compliance with financial regulations. To address these challenges, this paper introduces a novel fraud detection framework that integrates adversarial training with explainable AI (XAI) techniques for real-time fraud detection in high-frequency transactions. Our approach enhances model resilience against adversarial attacks by utilizing generative adversarial networks (GANs) and adversarial retraining, allowing the model to recognize and counteract fraudulent attempts more effectively. Furthermore, the incorporation of SHapley Additive exPlanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) provides interpretable decision-making, ensuring transparency and regulatory compliance. The proposed method is validated using real-world financial transaction datasets, benchmarking its performance against conventional machine learning models such as random forests, gradient boosting machines, and recurrent neural networks (RNNs). Experimental results demonstrate that our model significantly enhances fraud detection rates while maintaining high precision and recall, effectively reducing false negatives. Additionally, the XAI components facilitate clear model interpretability, enabling financial institutions to trust and adopt AI-driven fraud detection systems. This research contributes to the advancement of secure and interpretable AI applications in financial technology, paving the way for robust fraud prevention strategies in high-risk financial environments.
Article Details
Issue
Section
This work is licensed under a Creative Commons Attribution 4.0 International License.
How to Cite
References
A. Mahmood, K. S. Jayanth, and M. K. Hasan, “Machine Learning for Financial Fraud Detection: A Systematic Review,” IEEE Transactions on Neural Networks and Learning Systems, vol. 32, no. 8, pp. 3456–3471, 2023.
L. Chen, H. Fang, and P. Wang, “Adversarial Machine Learning for Fraudulent Transaction Detection,” Journal of Financial Analytics, vol. 15, no. 4, pp. 250–265, 2022.
X. Zhang, Y. Liu, and J. Lin, “A Comparative Study of Random Forest, Gradient Boosting, and Deep Learning for Financial Fraud Detection,” Neurocomputing, vol. 502, pp. 89–102, 2023.
S. Gupta, R. Kumar, and T. Li, “Handling Data Imbalance in Fraud Detection Using Synthetic Oversampling Techniques,” Expert Systems with Applications, vol. 199, p. 117314, 2022.
A. Smith and B. Taylor, “Unsupervised Learning for Anomaly Detection in Financial Transactions,” Proceedings of the 2022 International Conference on Artificial Intelligence & Finance, pp. 204–215, 2022.
P. Shah and M. Patel, “Time-Series Analysis in High-Frequency Fraud Detection Using LSTMs,” ACM Transactions on Computational Finance, vol. 8, no. 3, pp. 112–128, 2022.
H. Wu and L. Sun, “Graph-Based Semi-Supervised Learning for Fraud Detection in Digital Payments,” Pattern Recognition Letters, vol. 165, pp. 39–48, 2023.
J. Torres, M. Rivera, and A. Clarke, “Adversarial Training for Financial Fraud Detection: A Survey,” IEEE Transactions on Information Forensics and Security, vol. 18, pp. 58–72, 2023.
F. Liu, Y. Li, and G. Yang, “Explainable AI in Financial Fraud Detection: A Case Study Using SHAP and LIME,” Journal of Artificial Intelligence Research, vol. 75, pp. 310–327, 2023.
M. Wang and X. Zhao, “A Survey on Adversarial Attacks and Defenses in Deep Learning,” Neural Computing and Applications, vol. 35, no. 10, pp. 12245–12267, 2023.
C. Zhu, H. Zhang, and T. Lee, “Defending Against Adversarial Attacks in Financial AI Systems,” IEEE Security & Privacy, vol. 20, no. 1, pp. 22–31, 2023.
R. K. Gupta and P. Bansal, “Generative Adversarial Networks for Simulating Financial Fraud Patterns,” Proceedings of the 2023 International Conference on Machine Learning and Finance, pp. 145–160, 2023.
B. Tran and C. Liu, “Evaluating the Effectiveness of FGSM and PGD Attacks on Fraud Detection Models,” Journal of Cybersecurity and AI, vol. 18, no. 2, pp. 67–83, 2023.
A. Singh, “Enhancing Fraud Detection Models Using Adversarial Learning and Deep Reinforcement Learning,” IEEE Transactions on Computational Intelligence and AI in Finance, vol. 3, no. 1, pp. 99–114, 2023.
X. Li, “Interpretable Financial Fraud Detection with SHAP and LIME,” ACM Transactions on Explainable AI, vol. 5, no. 4, pp. 219–234, 2023.
D. Martinez and P. Ghosh, “Explainability in AI-Powered Risk Management Systems,” Journal of Computational Finance, vol. 10, no. 2, pp. 55–70, 2023.
M. Chen and Y. Huang, “Causal AI for Enhancing Transparency in Fraud Detection,” Expert Systems with Applications, vol. 195, p. 116943, 2023.
R. Thomas and L. Cooper, “Local Interpretability of Black-Box AI Models for Credit Card Fraud Detection,” Neural Processing Letters, vol. 58, no. 2, pp. 345–361, 2023.
H. Zhou and A. Kim, “Accelerating Fraud Detection with Edge AI and FPGA-Based Processing,” IEEE Transactions on Emerging Topics in Computational Intelligence, vol. 7, no. 1, pp. 122–135, 2023.
Ribeiro, M. T., Singh, S., & Guestrin, C. (2023). Beyond LIME: Counterfactual explanations for machine learning models in finance. Journal of Financial Data Science, 5(1), 45–62. https://doi.org/10.3905/jfds.2023.1.145
Lundberg, S. M., Erion, G. G., & Lee, S.-I. (2024). Hierarchical SHAP: Improved interpretability of complex models via hierarchical feature attribution. Machine Learning, 113(2), 359–389. https://doi.org/10.1007/s10994-023-06218-4