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{201488,
        author = {Pandi Durai K and Mrs.Anitha E M.E and Jeevanatham K and Sanjay M and Ganesan M},
        title = {A Data-Driven Approach to UPI Fraud Detection Using Machine Learning Techniques},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {3914-3918},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=201488},
        abstract = {With the rapid growth of digital payment systems, Unified Payments Interface (UPI) platforms have become a primary mode of financial transactions. However, this widespread adoption has also led to a significant rise in fraudulent activities, posing serious risks to users and financial institutions. This paper presents a machine learning-based approach for detecting fraudulent UPI transactions using a Random Forest classification model.
The proposed system analyses transaction-related features such as transaction amount, sender and receiver balances before and after the transaction, to identify suspicious patterns. A Random Forest algorithm is employed due to its robustness, ability to handle non-linear data, and effectiveness in reducing overfitting. The model is trained on a labelled dataset of financial transactions and achieves reliable performance in distinguishing between legitimate and fraudulent activities.
To enhance usability, the model is integrated into a user-friendly web application developed using Streamlit, allowing real-time prediction of transaction authenticity. The system provides quick and accurate results, making it suitable for practical deployment in digital payment platforms. Experimental results demonstrate that the proposed approach can significantly improve fraud detection accuracy while maintaining low false positive rates.
This work highlights the potential of machine learning techniques in strengthening the security of digital payment systems and offers a scalable solution for real-time fraud detection in UPI-based transactions.},
        keywords = {UPI Fraud Detection, Machine Learning, Random Forest, Digital Payment Security, Streamlit, Real-Time Prediction, Financial Fraud},
        month = {May},
        }