Loan Default Prediction Using Neural Networks

This project demonstrates how to predict loan default using an Artificial Neural Network (ANN) implemented in Python with TensorFlow and Keras.

📁 Project Structure

• loan_default_prediction.ipynb – Jupyter Notebook with model training, evaluation, and visualization.
• train.csv – Training dataset.
• test.csv – Test dataset.
• loan_predictions.csv – Example of model prediction outputs.

🔧 Techniques Used

• Data preprocessing with Pandas and Scikit-learn
• Standardization of numerical features
• One-hot encoding of categorical features
• Dropout and L2 Regularization to reduce overfitting
• Early stopping to optimize training duration
• Model evaluation using accuracy, loss, and confusion matrix

📊 Visualization

Training and validation loss over epochs were plotted to monitor learning progress.

🚀 How to Run

1. Clone the repository:

   git clone https://github.com/Mekusgood/loan-default-prediction.git