Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{201020,
        author = {Dr. Zaker Ul Oman and G Swathy and Kanishka Soudai},
        title = {AI-Driven Real-Time Fraud Detection in High-Volume UPI Transactions: Evaluating Instant Intervention Engines},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {2986-2996},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=201020},
        abstract = {The rapid growth of Unified Payments Interface (UPI) transactions in India has significantly accelerated digital financial inclusion However, this expansion has also increased vulnerability to sophisticated fraud risk. Traditional rule-based fraud detection systems are increasingly ineffective, as they struggle to handle large transaction volumes, evolving fraud patterns, and often generate high false positives with delayed detection.
This study proposes an advanced AI-driven intervention engine designed for real-time fraud detection in high-volume UPI environments. The system is designed to address critical challenges such as scalability and low-latency processing enabling efficient handling of billions of transactions. It adopts a multi-algorithm ensemble approach by integrating models such as Random Forest, XGBoost, and Neural Networks to enhance detection accuracy and effectively differentiate between legitimate and fraudulent activities.
A key contribution of this research is the incorporation of Explainable AI (XAI) techniques, including SHAP and LIME, which improve transparency and interpretability of automated decisions. These techniques help build trust among financial institutions and users by providing clear explanations for flagged transactions.Furthermore, the system utilizes behavioral and contextual analysis by examining transaction patterns, geographical locations, and device identifiers. This enables the detection of complex anomalies that traditional systems may overlook.
The proposed framework is aligned with regulatory standards, particularly the guidelines of the Reserve Bank of India (RBI) on digital security and data privacy, ensuring compliance and real-world applicability. Experimental results on large-scale datasets demonstrate that the model significantly reduces fraud-related losses and enhances the overall resilience of digital payment systems.
Overall, this research presents a scalable, intelligent, and secure solution aimed at strengthening trust and integrity within India’s UPI ecosystem.},
        keywords = {AI Fraud Detection, UPI Security, Real-Time Analytics, Explainable AI, Behavioral Analysis, Ensemble Learning, Financial Cybersecurity},
        month = {May},
        }