FraudEX-Smart ML Fraud Detection

Financial fraud in electronic payment systems has become a major challenge due to the rapid growth of digital banking, online transactions, and mobile payment platforms.Traditional rule-based fraud detection systems are unable to efficiently detect complex and evolving fraud patterns.This research presents FraudEx -Smart ML Fraud Detection, a machine learning-based system designed to identify fraudulent financial transactions with improved accuracy and reliability.The system utilizes supervised machine learning algorithms such as Logistic Regression, Decision Tree, and Random Forest to classify transactions as legitimate or fraudulent.Data preprocessing techniques including missing value handling, feature scaling, and class imbalance management were applied to enhance model performance.The models were evaluated using Accuracy, Precision, Recall, F1-Score, Confusion Matrix, and AUC-ROC metrics.Experimental results indicate that the Random Forest model achieved superior performance with improved fraud detection capability and reduced false positives.The proposed system demonstrates scalability, adaptability, and suitability for real-time fraud detection applications.