ARCHIVES
VOL. 10, ISSUE 6 (2025)
AI-Based secure monitoring and fraud detection for UPI payment transactions
Authors
Anoop Sharma
Abstract
The Unified Payments
Interface (UPI) has fundamentally transformed the digital payment landscape in
India, enabling billions of transactions every month across urban and rural
populations alike. However, this explosive growth has simultaneously attracted
a surge in fraudulent activities, including phishing, SIM swapping, social
engineering, and account takeover attacks. Traditional rule-based fraud
detection systems have proven inadequate against the dynamic and evolving
nature of modern payment fraud. This paper proposes and examines an Artificial
Intelligence (AI)-based framework for secure real-time monitoring and fraud
detection specifically designed for UPI payment transactions in the Indian
context. The proposed framework integrates machine learning algorithms
including Random Forest, Gradient Boosting, and Long Short-Term Memory (LSTM)
networks to analyze transaction patterns, detect anomalies, and flag suspicious
activities with high accuracy and low false-positive rates. The paper also
discusses data privacy considerations under India's Digital Personal Data
Protection Act, 2023, the role of the National Payments Corporation of India
(NPCI) in fraud governance, and the challenges unique to India's diverse
digital payment ecosystem. Experimental results and comparative analysis
suggest that AI-driven models significantly outperform traditional systems in
detection speed, accuracy, and adaptability. The paper further outlines future
research directions including federated learning for privacy-preserving fraud
detection and the integration of explainable AI (XAI) for regulatory
transparency.
Download
Pages:57-60
How to cite this article:
Anoop Sharma "AI-Based secure monitoring and fraud detection for UPI payment transactions". International Journal of Academic Research and Development, Vol 10, Issue 6, 2025, Pages 57-60
Download Author Certificate
Please enter the email address corresponding to this article submission to download your certificate.