AI-Driven Adaptive Fraud Detection Framework for Secure Online Financial Transactions

M. Sadiq Valli Khan, Mrs. B. Shireesha

Fraud Detection; Random Forest; Machine Learning; Django; Online Transactions.

Online financial fraud continues to impose substantial operational and reputational burdens on digital banking and e-commerce platforms. This manuscript presents an AI-driven fraud detection framework that combines a Random Forest ensemble classifier with a Django-based web application for real-time transaction screening. The proposed system evaluates behavioral and contextual signals, including transaction amount, temporal occurrence, geographic deviation from the registered profile, device novelty, location novelty, and transaction type, to estimate fraud risk. A synthetic dataset containing 10,000 transactions is generated to emulate realistic fraud patterns and to support controlled model training and validation. Experimental analysis indicates classification accuracy above 90%, strong precision–recall trade-offs, and an ROC-AUC exceeding 0.92 on held-out samples. The system further exposes feature importance values to support transparent administrative review and auditability. The resulting framework demonstrates that interpretable ensemble learning can deliver scalable, cost-effective, and deployment-ready fraud analytics for modern online transaction environments.

[1] L. Breiman, “Random Forests,” Machine Learning, Vol. 45, No. 1, 2001, pp. 5–32.
[2] A. Dal Pozzolo, O. Caelen, R. A. Johnson, and G. Bontempi, “Calibrating Probability with Undersampling for Unbalanced Classification,” IEEE Symposium Series on Computational Intelligence, 2015, pp. 159–166.
[3] S. Bhattacharyya, S. Jha, K. Tharakunnel, and J. C. Westland, “Data Mining for Credit Card Fraud: A Comparative Study,” Decision Support Systems, Vol. 50, No. 3, 2011, pp. 602–613.
[4] T. Chen and C. Guestrin, “XGBoost: A Scalable Tree Boosting System,” ACM SIGKDD, 2016, pp. 785–794.
[5] N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, “SMOTE: Synthetic Minority Over-sampling Technique,” Journal of Artificial Intelligence Research, Vol. 16, 2002, pp. 321–357.
[6] F. Pedregosa et al., “Scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research, Vol. 12, 2011, pp. 2825–2830.
[7] A. D. Pozzolo, G. Boracchi, O. Caelen, C. Alippi, and G. Bontempi, “Credit Card Fraud Detection: A Realistic Modeling and a Novel Learning Strategy,” IEEE Transactions on Neural Networks and Learning Systems, Vol. 29, No. 8, 2018, pp. 3784–3797.