About This Project

Machine Learning Based Credit Card Fraud Detection System

Project Overview

This project uses Machine Learning algorithms to identify fraudulent credit card transactions. Multiple models were trained and compared using various performance metrics.

The final model selected was XGBoost due to its superior fraud detection performance, excellent ROC-AUC score, and minimal overfitting.

Dataset

Credit Card Transaction Dataset with fraud and legitimate records.

Machine Learning

9 machine learning algorithms were trained and evaluated.

Best Model

XGBoost achieved the best performance among all tested models.

Technologies Used

Python Flask Pandas NumPy Scikit-Learn XGBoost Bootstrap 5 Machine Learning