AI-Driven Financial Auditing in Modern Banking: An Advanced Data-Driven Machine Learning Framework for Predictive Financial Intelligence, Real-Time Fraud Detection, and Automated Risk Monitoring
The rapid expansion of digital banking, real-time payment systems, and data-intensive financial services has increased the complexity of financial auditing and created new challenges related to fraud detection, financial-risk assessment, and continuous transaction monitoring. This study proposes an AI-driven financial auditing framework for modern banking that integrates predictive financial intelligence, real-time fraud detection, and automated risk monitoring within a unified data-driven machine learning architecture. A structured banking dataset containing 185,000 financial transactions from 12,500 anonymized customer accounts was developed from transactional, behavioural, account, authentication, credit, and risk-monitoring records collected over a three-year observation period. The dataset included 42 predictive variables, including transaction amount, transaction velocity, account balance variation, payment frequency, beneficiary novelty, failed-login intensity, device inconsistency, unusual access location, credit exposure, cash-flow volatility, historical fraud frequency, repayment behaviour, and audit-risk indicators. Following data cleaning, missing-value treatment, normalization, categorical encoding, class balancing, and feature selection, the dataset was divided into 70% training, 15% validation, and 15% testing subsets. Logistic Regression, Support Vector Machine, Random Forest, Artificial Neural Network, and XGBoost models were comparatively evaluated using accuracy, precision, recall, F1-score, ROC-AUC, and false-alert rate. Experimental results demonstrated that XGBoost achieved the strongest overall performance, obtaining 97.2% accuracy, 96.8% precision, 96.5% recall, 96.6% F1-score, and a ROC-AUC of 0.989, compared with 86.4% accuracy for Logistic Regression, 90.7% for SVM, 94.3% for Random Forest, and 95.8% for the neural network. The proposed auditing framework reduced false-positive audit alerts by 31.6%, improved suspicious-transaction detection by 24.8%, decreased average risk-identification time by 38.7%, and increased automated high-risk account recognition from 81.5% to 96.9%. Predictive financial analytics further achieved 94.6% accuracy in early financial-risk classification, enabling proactive identification of abnormal liquidity patterns, credit deterioration, and emerging transaction anomalies. Overall, the proposed framework demonstrates that machine learning can substantially improve audit accuracy, operational efficiency, fraud intelligence, and continuous financial-risk surveillance, providing a scalable foundation for intelligent and automated auditing in modern banking environments. References Adewale, T. T., Olorunyomi, T. D., & Odonkor, T. N. (2023). Big data-driven financial analysis: A new paradigm for strategic insights and decision-making. Journal of Financial Innovation and Analytics, 1(1), 1–15. Ahirrao, Y. S., Ansari, I., Azim, K. S., Bhujel, K., & Panchal, S. S. (2025). AI-powered financial strategy: Transforming business decision-making through predictive analytics. The USA Journals TAJET, 7(9), 126–151. Ahmad, B., Ali, S., Muneer, M., Fatima, A., & Abbas, N. (2025). Wind energy integration into the SAARC region: A comprehensive review of optimal wind sites for a sustainable super smart grid. Wind Energy, 3(2). Ahmad, B., Jillani, S. A., Rafique, M., & Soomro, A. A. (2026, January). Design and monitoring of a tree-inspired vertical axis wind turbine for efficient urban wind utilization. In 2026 1st International Conference on Innovations in Information and Communication Technologies (IICT) (pp. 1–6). IEEE. Ahmad, B., Majeed, M. K., Soomro, A. A., & Jillani, S. A. (2026, January). Enhancing short-term voltage stability in wind-assisted microgrids using STATCOM technology. In 2026 1st International Conference on Innovations in Information and Communication Technologies (IICT) (pp. 1–6). IEEE. Aljunaid, S. K., Almheiri, S. J., Dawood, H., & Khan, M. A. (2025). Secure and transparent banking: Explainable AI-driven federated learning model for financial fraud detection. Journal of Risk and Financial Management, 18(4), 179. Ashraf, S., & Khalid, N. (2024). Transforming financial services: AI-driven predictive analytics for smart operations and risk assessment. Bandi, V. D. V. K. (2024). AI-driven predictive risk modeling architectures for financial systems. International Journal of Finance, 37(3), 54–78. Beauty, A. M. (2025). Explainable AI in data-driven finance: Balancing algorithmic transparency with operational optimization demands. International Journal of Advanced Research Publication and Reviews, 2(6), 125–149. Celestin, M., & Mishra, A. K. (2025). AI-driven financial analytics: Enhancing forecast accuracy, risk management, and decision-making in corporate finance. Janajyoti Journal, 3(1), 1–27. Daryan, A., Kamyabi, N., Mehraban, S., & Khosravi, T. (2026). AI-driven fraud detection in financial statements: A comparative study of machine learning models. International Journal of Business Management and Entrepreneurship, 5(1), 24–40. Dorsey, J. (2025). AI-powered predictive analytics for risk management and fraud detection in modern banking (SSRN No. 7043898). SSRN. Elumilade, O. O., Ogundeji, I. A., Achumie, G. O., Omokhoa, H. E., & Omowole, B. M. (2021). Enhancing fraud detection and forensic auditing through data-driven techniques for financial integrity and security. Journal of Advanced Education and Sciences, 1(2), 55–63. Garud, S. (2025). AI-driven risk management in financial services from theory to practice in smart education. In Smart education and sustainable learning environments in smart cities (pp. 77–92). IGI Global Scientific Publishing. Green, A. (2025). AI-driven financial intelligence systems: A new era of risk detection and strategic analysis. Center for Open Science. Ismaeil, M. K. A. (2024). Harnessing AI for next-generation financial fraud detection: A...