Unified Payments Interface Fraud Detection Using Machine Learning

P. Sirisha, S. Jaheda, G. Afra Tahaseen, K. Divya Madhuri, P. Mohammad Arshad

2025

Abstract

Digital transaction fraud has become more common due to the widespread use of the Unified Payments Interface (UPI). Our proposed fraud detection system employs six machine learning algorithms to address this issue: XGBClassifier, Decision Tree, Random Forest, and Gradient Boosting Machines (GBMs). A logical framework for classifying transactions is provided by the Decision Tree algorithm. By enhancing accuracy and resilience, Random Forest successfully detects fraud. By combining weak learners, GBMs are able to identify changing fraud trends over time and capture intricate fraud patterns. training the model to converge efficiently. When it comes to classification jobs, XGBClassifier is a formidable gradient boosting method. Quick, it deals with missing values, and it stops overfitting. Enhancing UPI security, this multi-algorithm technique processes UPI transactions securely and accurately, differentiating between fraudulent and authentic ones. We are on the cusp of seeing this paradigm implemented in actual financial systems.

Paper Citation

in Harvard Style

Sirisha P., Jaheda S., Tahaseen G., Madhuri K. and Arshad P. (2025). Unified Payments Interface Fraud Detection Using Machine Learning. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 420-426. DOI: 10.5220/0013930800004919

in Bibtex Style

@conference{icrdicct`2525,
author={P. Sirisha and S. Jaheda and G. Tahaseen and K. Madhuri and P. Arshad},
title={Unified Payments Interface Fraud Detection Using Machine Learning},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={420-426},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013930800004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - Unified Payments Interface Fraud Detection Using Machine Learning
SN - 978-989-758-777-1
AU - Sirisha P.
AU - Jaheda S.
AU - Tahaseen G.
AU - Madhuri K.
AU - Arshad P.
PY - 2025
SP - 420
EP - 426
DO - 10.5220/0013930800004919
PB - SciTePress