Author: Camille Maslin
Contact:
Description: This project implements a high-performance credit card fraud detection system using advanced machine learning techniques. The model achieves 99.97% accuracy on real transaction data.
Source: Kaggle - Credit Card Fraud Detection Dataset 2023
Size: 57,000+ transactions
Features:
- Transaction amount
- Time of transaction
- 28 anonymized features (V1-V28)
- Target: Binary classification (Fraud/Non-Fraud)
Data Quality:
- No missing values
- Preprocessed and anonymized for privacy
- Standardized numerical features
- Reflects real-world transaction patterns
- Comprehensive data exploration
- Advanced feature engineering
- Robust data quality checks
- Interactive correlation matrices
- Distribution analysis
- Pattern recognition plots
- 99.97% accuracy rate
- Only 18-19 errors per 57,000 transactions
- SMOTE implementation for class balancing
- Cross-validation scores: [0.9996 - 0.9997]
- Balanced precision and recall
- Minimal false positives/negatives
- 🐍 Python 3.12
- 📝 Scikit-learn
- 📊 Pandas & NumPy
- 🔄 Matplotlib & Seaborn
- 🔄 SMOTE for imbalance handling
- Clone the repository:
$ git clone https://github.com/camille-maslin/SecureCard-AI.git
$ cd SecureCard-AI- Create a virtual environment and activate it:
$ python3 -m venv venv
$ source venv/bin/activate # Linux/MacOS
$ .\venv\Scripts\activate # Windows- Install dependencies:
$ pip install -r requirements.txt- Run the Jupyter Notebook:
$ jupyter notebook-
Open
SecureCard-AI.ipynbin your Jupyter environment. -
Follow the instructions and run each cell to:
- Load data
- Analyze and preprocess the dataset
- Train the fraud detection model
- Evaluate performance and visualize results
This project is licensed under the MIT License - see the LICENSE file for details.
Contributions are welcome! Please submit a pull request or open an issue for suggestions or bug reports.
- Kaggle for the dataset.
- Open-source libraries and contributors for tools used.