Credit Card Fraud Detection

This project focuses on detecting fraudulent credit card transactions using machine learning techniques. The dataset used is highly imbalanced, with a small percentage of transactions classified as fraud. To address this imbalance, we apply Synthetic Minority Over-sampling Technique (SMOTE) to balance the dataset before training models like Logistic Regression and Decision Tree.

Project Overview

Fraud detection is critical for financial institutions to prevent losses and protect customers. In this project, we:

• Explore and analyze the credit card transactions dataset.
• Apply SMOTE to handle the imbalanced data problem.
• Train and evaluate Logistic Regression and Decision Tree models.
• Use evaluation metrics like Confusion Matrix, ROC-AUC score, and Classification Report to measure model performance.