Fraud Detection System - Advanced Version

Overview

Detect fraudulent transactions using machine learning models with SMOTE handling for class imbalance. Models included: Logistic Regression, Random Forest, XGBoost, SVM.

How to Run

1. Install dependencies: pip install pandas scikit-learn xgboost imbalanced-learn matplotlib seaborn
2. Run the model: python fraud_model.py
3. Check dashboard_fraud.png for the confusion matrix.

Next Steps

• Hyperparameter tuning
• Improve dashboard with feature importance and ROC curves