Authors :
Poomathi M.; Mirunalini A.
Volume/Issue :
Volume 11 - 2026, Issue 10 - October
DOI :
https://doi.org/10.38124/ijisrt/26Oct663
Note : A published paper may take 4-5
working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and
ResearchGate.
Abstract :
The rapid growth of Unified Payments Interface (UPI) transactions has increased the need for effective methods to detect and prevent digital payment fraud. This project proposes a machine learning-based UPI fraud detection system that analyzes transaction details and identifies patterns that may indicate suspicious activity. Features such as transaction amount, time, frequency, recipient information, device details, and transaction history are used to assess each payment.The system applies machine learning to classify transactions and generate an AI-based risk score. It provides explanations for flagged transactions, such as unusually large payments, repeated transfers within a short period, or payments to unfamiliar recipients. Based on configurable risk thresholds, the system can approve low-risk transactions, send medium-risk transactions for review, and block high-risk transactions. A web-based dashboard displays transaction records, risk scores, alerts, explanations, and decision status for users and administrators.The proposed system aims to support faster fraud monitoring and provide a structured process for handling suspicious payments. Its performance depends on the quality and representativeness of the transaction data used for training and evaluation. The system is intended as a fraud prevention support tool, with final deployment and transaction blocking dependent on integration with authorized payment services
Keywords :
UPI, Machine Learning, Fraud Detection, AI-Based Risk Scoring, Transaction Blocking, Digital Payments, Anomaly Detection.
References :
- Sethi, Bhramanand, Sarvednya Mhatre, Sachin Yadav, Siuli Das, and Vaishali Jadhav. “Machine Learning-Based UPI Fraud Detection: A Comprehensive Approach Using Random Forest.” Proceedings of ICAIEHS 2025, 2025. https://doi.org/10.2991/978-94-6463-852-3_29
- Sree, Mangali Vijaya, Panyam Bhuvaneshwari, Karne Nandini, Koppula Ratna Phoebe Amulya, Palla Swetha, and B. Chandrakala. “Enhanced Detection of UPI Frauds Using Advanced Machine Learning Techniques for Secure Digital Transactions.” Proceedings of ICITSM 2025, 2025. https://doi.org/10.4108/eai.28-4-2025.2357830
- “A Novice Approach for UPI Fraud Detection by Using Machine Learning Techniques.” 2025 IEEE 7th International Conference on Computing, Communication and Automation (ICCCA), 2025. https://doi.org/10.1109/ICCCA66364.2025.11325722
- “AI-Driven Detection Mechanism for UPI Fraud and QR Code Tampering.” 2025 6th International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV), 2025. https://doi.org/10.1109/ICICV64824.2025.11085656
- “FedUPI: Federated Learning Empowered Detection of Fraudulent UPI Transactions.” 2025 5th International Conference on Soft Computing for Security Applications (ICSCSA), 2025. https://doi.org/10.1109/ICSCSA66339.2025.11170880
- “Secure UPI: Machine Learning-Driven Fraud Detection System for UPI Transactions.” 2024 2nd International Conference on Disruptive Technologies (ICDT), 2024. https://doi.org/10.1109/ICDT61202.2024.10489682
- “Experimental Evaluation of Unified Payment Interface (UPI) Fraud Detection System Using Elevated Deep Learning Methodology.” 2024 2nd International Conference on Emerging Trends in Engineering and Medical Sciences (ICETEMS), 2024. https://doi.org/10.1109/ICETEMS64039.2024.10965083
- “Fraud Fighters: How AI and ML are Revolutionizing UPI Security.” 2024 International Conference on Science Technology Engineering and Management (ICSTEM), 2024. https://doi.org/10.1109/ICSTEM61137.2024.10560740
- Sindhu, Jallapuram, and Vijaya Sree Swarupa. “UPI Fraud Detection Using Machine Learning Algorithms.” International Journal of Engineering Research and Science & Technology, Vol. 20, No. 4, pp. 57–67, 2024. Article link
- Anupriya, K., K. Avinash, S. Hariharan, R. Bhavesh, and R. R. Kumaran. “LLM-Powered UPI Transaction Monitoring and Fraud Detection.” 2024 International Conference on System, Computation, Automation and Networking (ICSCAN), 2024. https://doi.org/10.1109/ICSCAN62807.2024.10894012
The rapid growth of Unified Payments Interface (UPI) transactions has increased the need for effective methods to detect and prevent digital payment fraud. This project proposes a machine learning-based UPI fraud detection system that analyzes transaction details and identifies patterns that may indicate suspicious activity. Features such as transaction amount, time, frequency, recipient information, device details, and transaction history are used to assess each payment.The system applies machine learning to classify transactions and generate an AI-based risk score. It provides explanations for flagged transactions, such as unusually large payments, repeated transfers within a short period, or payments to unfamiliar recipients. Based on configurable risk thresholds, the system can approve low-risk transactions, send medium-risk transactions for review, and block high-risk transactions. A web-based dashboard displays transaction records, risk scores, alerts, explanations, and decision status for users and administrators.The proposed system aims to support faster fraud monitoring and provide a structured process for handling suspicious payments. Its performance depends on the quality and representativeness of the transaction data used for training and evaluation. The system is intended as a fraud prevention support tool, with final deployment and transaction blocking dependent on integration with authorized payment services
Keywords :
UPI, Machine Learning, Fraud Detection, AI-Based Risk Scoring, Transaction Blocking, Digital Payments, Anomaly Detection.
