FraudGuard ML: Advanced Payment Fraud Detection System

A sophisticated machine learning-based payment fraud detection system that leverages multiple algorithms (Random Forest, XGBoost, LightGBM, CatBoost) to identify suspicious transactions. Features 50,000 synthetic transactions, 8 data visualizations, and SMOTE for handling class imbalance. Perfect for financial institutions and data scientists working on fraud detection systems.

🚀 Features

• Dataset Generation: 50,000 synthetic transactions with 10 relevant features
• Multiple ML Models: Implementation of 4 powerful algorithms
  • Random Forest
  • XGBoost
  • LightGBM
  • CatBoost
• Advanced Visualizations: 8 different types of data visualizations
• Class Imbalance Handling: SMOTE technique implementation
• Comprehensive Evaluation: Multiple performance metrics

📊 Project Structure

fraud_detection/
├── data/               # Dataset directory
├── docs/              # Documentation and visualizations
├── models/            # Trained models
├── notebooks/         # Jupyter notebooks
├── src/              # Source code
└── tests/            # Unit tests

🛠️ Technical Stack

• Languages: Python 3.x
• ML Libraries:
  • Scikit-learn
  • XGBoost
  • LightGBM
  • CatBoost
• Data Processing:
  • Pandas
  • NumPy
• Visualization:
  • Matplotlib
  • Seaborn
  • Plotly
• Development:
  • Jupyter Notebooks
  • Git

🚀 Getting Started

1. Clone the repository

git clone https://github.com/Arindam-GitH/Payment-Fraud-Detection-.git
cd Payment-Fraud-Detection-

2. Set up virtual environment

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies

pip install -r requirements.txt

4. Run the analysis

python src/main.py

📈 Model Performance

Each model is evaluated using:

• Accuracy
• Precision
• Recall
• F1-Score
• ROC-AUC Score

📊 Visualizations

The project includes 8 different types of visualizations:

1. Transaction Amount Distribution
2. Fraud vs Non-Fraud Distribution
3. Correlation Heatmap
4. Transaction Time Analysis
5. Amount vs Fraud Box Plot
6. Customer Age Distribution
7. Device Type Distribution
8. Geographic Distribution

🤝 Contributing

Feel free to submit issues and enhancement requests!

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

👤 Author

Arindam Guha

• GitHub: @Arindam-GitH

---

⭐️ From Arindam-GitH