ML-Based Credit Scoring System

A full-stack machine learning web application that predicts loan eligibility using a trained ML model, a Flask backend API, and a Flutter web frontend.

---

🚀 Project Overview

This project is an end-to-end Credit Scoring System that evaluates loan eligibility based on customer financial and demographic data.
It includes:

• Machine Learning Model (trained using joblib)
• Flask Backend API
• SHAP Explainability showing top 5 feature contributions
• Flutter Web Frontend for user interaction
• Render Deployment for hosting the backend + frontend

---

🧠 Features

🔍 ML Model

• Trained classification model
• Predicts loan approved or loan rejected
• Exported using joblib

🧩 Explainability (SHAP)

The API returns the top 5 contributing features for every prediction.

🌐 API Backend (Flask)

• Predict loan eligibility
• Serve static Flutter web files
• CORS enabled
• Health check endpoint /healthz for Render

🎨 Flutter Web Frontend

• User-friendly form
• Sends data to the backend
• Displays loan approval + SHAP insights

---

🗂️ Repository Structure