🏦 Loan Approval Predictor

A professional Machine Learning system designed to predict bank loan approval decisions, featuring an interactive web interface built with Streamlit.

---

🎯 Project Objective

The goal of this project is to build an intelligent system capable of automatically predicting whether a loan application will be approved or rejected, based on:

• Demographic information (age)
• Financial data (annual income, requested loan amount, monthly deductions)
• Product type and decision process
• Historical banking decisions

This project simulates a real-world retail banking use case with production-oriented practices.

---

✨ Key Features

🤖 Machine Learning

• 6 machine learning algorithms implemented and compared
  • Decision Tree
  • Random Forest
  • Extra Trees
  • XGBoost
  • LightGBM
  • CatBoost
• Automated hyperparameter optimization
• Stratified cross-validation
• Comprehensive evaluation metrics:
  • Accuracy
  • Precision
  • Recall
  • F1-score
  • AUC-ROC

---

📊 Data Analysis & Feature Engineering

• Automated Exploratory Data Analysis (EDA)
• Advanced visualizations (20+ charts)
• Feature engineering:
  • Financial ratios
  • Feature interactions
• Outlier detection and data cleaning

---

🌐 Web Application (Streamlit)

• Interactive dashboard for data exploration
• Real-time loan approval prediction
• Dynamic visualizations using Plotly
• One-click model comparison

---

🛠️ Software Engineering

• Modular and maintainable architecture
• Unit testing with pytest
• Fully documented codebase
• Optimized VS Code configuration

---

📦 Installation

Prerequisites

• Python 3.12+
• pip
• Git (optional)

Quick Setup

# 1. Clone the repository
git clone https://github.com/malek-harbaoui/Loan-Approval-Predictor.git
cd Loan_Approval_Predictor

# 2. Create a virtual environment
python -m venv venv

# 3. Activate the environment
# Windows
venv\Scripts\activate
# Linux / macOS
source venv/bin/activate

# 4. Install dependencies
pip install -r requirements.txt

# 5. Verify installation
python -c "import pandas, sklearn, xgboost; print('Installation OK')"

---

🚀 Usage

📊 Full ML Pipeline

# 1. Exploratory Data Analysis
python scripts/run_eda.py

# 2. Data preprocessing & feature engineering
python scripts/run_preprocessing.py

# 3. Train machine learning models
python scripts/train_models.py

# 4. Generate performance reports
python scripts/generate_report.py

🌐 Launch the Web Application

streamlit run app.py

Then open your browser at: 👉 http://localhost:8501

🎯 Programmatic Usage

from src.data.data_loader import DataLoader
from src.models.boosting_models import XGBoostModel
from sklearn.model_selection import train_test_split

# Load data
loader = DataLoader()
df = loader.load_retail_data()

# Prepare features and target
X = df.drop("Décision Finale Binaire", axis=1)
y = df["Décision Finale Binaire"]

X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42
)

# Train model
model = XGBoostModel()
model.build_model(n_estimators=200, learning_rate=0.3)
model.fit(X_train, y_train)

# Predict
predictions = model.predict(X_test)
probabilities = model.predict_proba(X_test)

---

📁 Project Structure

Loan_Approval_Predictor/
│
├── app.py                    # 🌐 Streamlit application
├── requirements.txt          # 📦 Dependencies
├── config.yaml               # ⚙️ Global configuration
├── README.md                 # 📖 Documentation
│
├── data/
│   ├── raw/                  # Raw datasets
│   └── processed/            # Cleaned datasets
│
├── src/
│   ├── data/                 # Data processing
│   ├── models/               # ML models
│   ├── visualization/        # Charts & plots
│   └── evaluation/           # Metrics & evaluation
│
├── scripts/                  # Execution scripts
├── notebooks/                # Jupyter notebooks
├── tests/                    # Unit tests
├── models/                   # Saved models
└── reports/                  # Results & figures

---

📊 Expected Results

Model Performance

Model	Accuracy	Precision	Recall	F1-Score	AUC
XGBoost	0.924	0.918	0.931	0.924	0.957
LightGBM	0.921	0.915	0.928	0.921	0.954
CatBoost	0.918	0.912	0.925	0.918	0.951
Random Forest	0.915	0.909	0.922	0.915	0.948

🔍 Most Important Features

1. Retenus Mensuel - 35.2%
2. Revenus Annuel - 28.7%
3. Montant Sollicité - 18.4%
4. Age - 9.3%
5. Type CANEVAS - 4.8%

---

🎨 Streamlit Interface

Available Pages

1. 🏠 Home

   • Dataset overview
   • Key statistics
   • Model performance indicators

2. 📊 Data Exploration

   • Feature distributions
   • Bivariate analysis
   • Correlation matrix
   • Interactive visualizations

3. 🎯 Prediction

   • User input form
   • Real-time loan approval prediction
   • Decision analysis
   • Explanation of influencing factors

4. 📈 Model Results

   • Model performance comparison
   • Performance charts
   • Detailed evaluation metrics
   • Result visualizations

---

🛠️ Technologies Used

Core ML

• scikit-learn
• XGBoost
• LightGBM
• CatBoost

Data Processing

• pandas
• numpy

Visualization

• matplotlib
• seaborn
• plotly

Web Application

• Streamlit
• Streamlit-Plotly

Development Tools

• pytest
• black
• flake8

---

🔧 Global Configuration (config.yaml)

project:
  name: "Loan Approval Predictor"
  version: "1.0.0"
  random_seed: 42

data:
  test_size: 0.2
  target_column: "Décision Finale Binaire"

models:
  xgboost:
    n_estimators: 200
    learning_rate: 0.3
    max_depth: 4

🔐 Environment Variables

Create a .env file at the project root:

DATA_PATH=data/raw
MODEL_PATH=models
RANDOM_STATE=42

---

📊 Evaluation Metrics

Le projet utilise plusieurs métriques pour évaluer les modèles :

• Accuracy – Overall correctness of predictions
• Precision – Proportion of correct positive predictions
• Recall - Ability to detect positive cases
• F1-Score - Harmonic mean of Precision and Recall
• AUC-ROC - Area under the ROC curve
• Specificity - Ability to detect negative cases

---

🤝 Contributing

Contributions are welcome! To contribute to this project:

1. Fork the repository
2. Create a new branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

---

👥 Auteur

• Malek Harbaoui - Main Developer

---

⭐ If this project helped you, please consider giving it a star!