• Dr C Nadhamuni Reddy
• Mr E D Pavan Kumar
• S Karishma
• Y Prashanthi
• M Varshini,D Shafi Qur Rehaman

As digital transactions have become more common, UPI (Unified Payments Interface) fraud has increased, making strong detection methods necessary. In order to detect fraudulent activity, this project offers a machine learning-based UPI fraud detection system that examines transaction patterns. Transaction information including amount, merchant category, frequency, and location anomalies make up the dataset. We handle missing values, encode category variables, and normalise numerical features as part of our preprocessing steps. A number of machine learning models are trained and assessed, such as Random Forest, Logistic Regression, Gradient Boosting, and XGBoost. The highest level of fraud detection accuracy is guaranteed by feature importance analysis. The top-performing model successfully separates fraudulent transactions from authentic ones by achieving high precision and recall. By reducing financial risks associated with digital payments and offering real-time fraud detection, this technology improves financial security.

Data preprocessing, feature engineering, random forest, xgboost, real-time detection, cybersecurity, risk mitigation, machine learning, digital payments, transaction analysis, anomaly detection, fraud prevention, financial security,UPI fraud detection.

"UPI FRAUD DETECTION USING MACHINE LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 4, page no.f490-f496, April-2025, Available :http://www.jetir.org/papers/JETIR2504567.pdf

"UPI FRAUD DETECTION USING MACHINE LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 4, page no. ppf490-f496, April-2025, Available at : http://www.jetir.org/papers/JETIR2504567.pdf