Alfazril/credit-risk-prediction

🏦 Credit Risk Prediction System

Sistem prediksi risiko kredit menggunakan ensemble machine learning models.

🚀 Features

• Ensemble Model: Kombinasi dari XGBoost, LightGBM, CatBoost, dan LSTM
• Real-time Prediction: Prediksi risiko kredit secara real-time
• User-friendly Interface: Antarmuka yang mudah digunakan dengan Gradio
• Comprehensive Analysis: Analisis mendalam dengan rekomendasi

📊 Models Used

1. XGBoost: Gradient boosting model
2. LightGBM: Light gradient boosting model
3. CatBoost: Categorical boosting model
4. LSTM: Long Short-Term Memory neural network
5. Improved LSTM: Enhanced LSTM model

🔧 Input Features

• Personal Information: Age, Income, Employment Length
• Loan Information: Loan Amount, Interest Rate, Loan-to-Income Ratio
• Credit History: Credit History Length, Number of Credit Lines, Credit Utilization
• Payment History: Payment History Status, Delinquencies

📈 Output

• Risk Probability (0-100%)
• Risk Level (Low/Medium/High)
• Individual Model Predictions
• Recommendations

🛠️ Technology Stack

• Python 3.9+
• Gradio
• TensorFlow/Keras
• XGBoost, LightGBM, CatBoost
• Pandas, NumPy, Scikit-learn

📝 License

Apache License 2.0