• Jakka Venkata Vara Sidhu Naidu

This paper introduces a thorough machine learning methodology for identifying fraudulent activities within the Unified Payments Interface (UPI) ecosystem. With the increasing digitalization of financial transactions, UPI has become a key payment mechanism in many regions. However, the rapid adoption of UPI has led to sophisticated fraudulent schemes. In this work, we leverage a rich dataset obtained from Kaggle, which contains both legitimate and fraudulent transaction records. A range of machine learning algorithms, such as Logistic Regression, Decision Trees, Random Forest, Neural Networks, and gradient boosting methods, are evaluated. We also employ extensive feature engineering techniques and anomaly detection methods to improve detection accuracy. Experimental findings indicate that ensemble methods substantially improve detection efficacy. Future research avenues encompass the amalgamation of deep learning (DL) frameworks and creation of of adaptive, real-time detection systems.

UPI fraud detection, machine learning, anomaly detection, ensemble learning, digital payments.

"Machine Learning-Based UPI Fraud Detection System", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 3, page no.g692-g694, March-2025, Available :http://www.jetir.org/papers/JETIR2503683.pdf

"Machine Learning-Based UPI Fraud Detection System", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 3, page no. ppg692-g694, March-2025, Available at : http://www.jetir.org/papers/JETIR2503683.pdf