Paysure: Smart UPI Payments with Real-Time Fraud Detection.

The PaySure project provides a comprehensive solution for real-time fraud detection in UPI-based digital transactions. By employing advanced machine learning techniques, particularly a highly optimized CatBoost gradient boosting classifier, the system can effectively identify anomalous and potentially fraudulent activities with high accuracy. The integration of real-time monitoring ensures that users are promptly alerted to suspicious transactions, thereby reducing financial risk and improving overall trust in digital payment systems. The adoption of the Agile development methodology allowed for iterative improvements, quick feedback incorporation, and progressive enhancement of both the ML models and the system’s user interface. This approach ensured that the system could adapt to changing fraud patterns and evolving user requirements. Moreover, the use of CRISP-DM for structured data mining and model development ensured that the solution was data-driven, reliable, and scalable. In summary, PaySure not only strengthens the security infrastructure of UPI payment systems but also serves as a flexible and extensible framework that can be applied to other digital payment platforms. The project highlights the importance of combining AI techniques with best software development practices to achieve effective, real-time fraud detection.

[1]           M. Anjali, M. Ajay, M. Saiteja, and B. O. Yadav, “Paysafe ai: Intelligent fraud detection for upi transactions using machine learning,” International Journal of Innovative Research in Computer and Communication Engineering, pp. 112–118, 2025. [2]           R. A. Kumar, S. Ishrat, M. D. Prasad, P. A. Siddiq, and N. C. H. Shankar, “Ai-driven detection mechanism for upi fraud and qr code tampering,” in IEEE International Conference on Computational Intelligence and Data Science, pp. 330–337, 2025. [3]           D. M. Rao, T. R. S. Varun, T. V. Gupta, and D. M. Kumar, “Comparative analysis of upi fraud detection using ensemble learning,” International Journal of Engineering and Advanced Technology, pp. 452–458, 2025. [4]           Rethisha.R and S. Cyindia.H, “Leveraging machine learning techniques for real-time detection of upi fraud,” International Research Journal of Modernization in Engineering Technology and Science, pp. 502–509, 2025. [5]           G. H. Trinity and N. Sharma, “Unified payments interface (upi): Fraud and prevention strategies,” Journal of Banking and Financial Technology, pp. 88–94, 2025. [6]           N. Kumar, K. Hemavathi, S. G, K. Sravani, T. Aprna, and M. Deepthi, “Enhancing upi security through deep learning-based fraud detection,” International Journal of Research Publication and Reviews, pp. 212–219, 2025. [7]           S. Bharath, G. L. V. Prasad, V.Sujatha, S.Hemajothi, D. Mani, and N. G. Merlin, “Hmlm: An intelligent artificial intelligence assisted strategy to identify upi frauds based on hybrid markov learning methodology,” 2024. [8]           V. Gupta, S. Sharma, S. Pathak, and S. Nimkar, “Upi based financial fraud detection using deep learning approach,” 2024. [9]           M. Tamilselvi, J. Giri, and D. Sheela, “Experimental evaluation unified payment interface (upi) fraud de tection system using elevated deep learning methodology,” 2024. [10]        R. Rani, A. Alam, and A. Javed, “Secure upi: Machine learning-driven fraud detection system for upi transactions,” 2024.