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{198045,
        author = {Kshitija Patil and Nikita kisan shinde},
        title = {UPI Fraud Detection Using AI},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {10672-10677},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=198045},
        abstract = {The swift expansion of online monetary interfaces, with a specific emphasis on the Unified Payments Interface (UPI), has fundamentally altered the landscape of financial exchanges, facilitating immediate, cash-free, and fluid fund movements. Conversely, this surge in transactional volume has concurrently triggered a marked escalation in deceptive practices, ranging from credential harvesting and identity cloning to illicit account access and unapproved monetary shifts. Legacy systems dependent on static rules frequently prove insufficient when tackling dynamic fraud schemes as well as skewed data distributions within transaction logs. [1], [3] In response, this research outlines an AI-centric architecture for identifying illicit activities within online payment infrastructures, utilizing cutting-edge Machine Learning methodologies. The suggested algorithm scrutinizes attributes related to time, user behavior, and transaction details derived from past payment records to precisely categorize activities as either authentic or malicious. Various supervised learning models, encompassing Gradient Boosting, Support Vector Machines, Logistic Regression, and Random Forest, were deployed and subjected to comparative analysis. [2], [4] To minimize erroneous alerts and boost efficacy, methods involving hyperparameter tuning and feature refinement were employed. Empirical findings suggest that AI-based approaches deliver superior metrics concerning accuracy, precision, recall, and F1-scores, validating their capability for instantaneous fraud identification. Ultimately, this framework strengthens the safety, dependability, and credibility of payment networks through early risk evaluation and prompt countermeasures against illicit actions.},
        keywords = {Artificial Intelligence, Digital Payment Systems, Fraud Detection, Machine Learning, UPI Security, Anomaly Detection, Financial Cybersecurity.},
        month = {April},
        }