Skip to content

Geetanjali-147/Sahay-Score

BranchesTags

Repository files navigation

🎯 SahayScore - AI-Powered Dual Credit Scoring System

SahayScore Banner React Vite TailwindCSS Node.js License

Empowering financial inclusion through AI-powered dual credit scoring for backward classes

Live Demo • Documentation • Quick Start • Features

📖 Overview

SahayScore is an innovative loan assessment platform developed for the National Backward Classes Finance & Development Corporation (NBCFDC). It revolutionizes the lending process by implementing a dual credit scoring system that evaluates both repayment capability and genuine financial need.

🎯 Mission

Transform the loan approval process from 7 days to just a few hours while maintaining fairness and reducing defaults through AI-powered scoring.

🌟 Key Innovation

Unlike traditional credit scoring systems that only assess repayment ability, SahayScore introduces a 2D scoring matrix that simultaneously evaluates:

  1. Repayment Score (0-500): Based on credit history, previous loans, and business income
  2. Need Score (0-500): Based on consumption proxies (electricity bills, mobile recharges, utility payments)

✨ Features

🚀 For Applicants

  • 📝 Smart Application Form

    • Real-time validation for Aadhar, phone, and loan amounts
    • Multi-step form with consumption-based need assessment
    • Document upload support (PDF, JPG, PNG up to 5MB)
    • Instant feedback and error handling

  • 📊 Instant Score Results

    • Composite score out of 1000
    • Dual score breakdown with visual indicators
    • Clear classification (Auto-Approve/Manual Review/Reject)
    • Next steps and loan terms for approved applications

  • 🔍 Application Dashboard

    • View all submitted applications
    • Search by name or application ID
    • Filter by status (Approved, Pending, Rejected)
    • Track application progress

🛡️ For Administrators

  • ⚙️ Admin Panel

    • Comprehensive application management
    • Advanced search and filtering
    • Approve/reject applications
    • Detailed scoring breakdown and classification

  • 📈 Analytics & Insights

    • Application statistics
    • Approval rate tracking
    • Score distribution analysis
    • Performance metrics

🎨 Design & UX

  • Modern Gradient UI with orange-green NBCFDC branding
  • Framer Motion animations for smooth interactions
  • Fully responsive design for mobile, tablet, and desktop
  • Accessibility compliant (ARIA labels, keyboard navigation, screen reader support)
  • Dark mode support (optional)

🏗️ Architecture

Tech Stack

Frontend

  • React 18.3.1 - UI library
  • Vite 5.4.20 - Build tool and dev server
  • Tailwind CSS 3.4.17 - Utility-first CSS framework
  • Framer Motion 12.23.22 - Animation library
  • Lucide React 0.545.0 - Icon library

Backend

  • Node.js - Runtime environment
  • Express.js - Web framework
  • In-memory storage - Application data (upgradeable to MongoDB/PostgreSQL)
  • RESTful API - Backend communication

Development Tools

  • PostCSS - CSS processing
  • Autoprefixer - CSS vendor prefixing
  • Concurrently - Run multiple commands

Project Structure

SahayScore/
├── backend/
│   ├── server.js              # Express API server
│   └── package.json           # Backend dependencies
├── src/
│   ├── app.jsx                # Main React component (1566 lines)
│   ├── constants.js           # Scoring weights & thresholds
│   ├── utils.js               # Validation & API utilities
│   ├── index.css              # Global styles
│   └── main.jsx               # React entry point
├── index.html                 # HTML template
├── package.json               # Frontend dependencies
├── vite.config.js             # Vite configuration
├── tailwind.config.js         # Tailwind configuration
├── postcss.config.js          # PostCSS configuration
├── CLASSIFICATION_MATRIX.md   # Detailed scoring logic
├── QUICK_START.md             # Setup guide
├── IMPROVEMENTS_SUMMARY.md    # Changelog
└── README.md                  # This file

🚀 Quick Start

Prerequisites

  • Node.js (v14 or higher)
  • npm or yarn
  • Modern web browser (Chrome, Firefox, Safari, Edge)

Installation

  1. Clone the repository

    git clone https://github.com/yourusername/SahayScore.git
cd SahayScore

  2. Install frontend dependencies

    npm install

  3. Install backend dependencies

    cd backend
npm install
cd ..

Running the Application

Option 1: Run Both Servers Together (Recommended)

npm run dev

This starts both frontend (port 5173) and backend (port 5000) concurrently.

Option 2: Run Separately

Terminal 1 - Backend:

cd backend
npm start

✅ Backend running at: http://localhost:5000

Terminal 2 - Frontend:

npm run dev

✅ Frontend running at: http://localhost:5173

Access the Application

Open your browser and navigate to:

📚 Documentation

Scoring System

Repayment Score (0-500)

Calculated based on:

  • Previous Loans: +40 points per loan (max 200)
  • Business Income: Higher income = higher score (max 150)
  • Base Score: 150 points

Formula:

repaymentScore = 150 + (previousLoans × 40) + (businessIncome / 1000) × 5

Need Score (0-500)

Calculated based on consumption proxies (lower consumption = higher need):

  • Electricity Bill: Lower bills indicate higher need
  • Mobile Recharge: Lower recharges indicate higher need
  • Utility Payments: Lower payments indicate higher need

Formula:

needScore = (200 - electricityBill/10) + (150 - mobileRecharge/5) + (150 - utilityPayments/10)

Composite Score (0-1000)

compositeScore = repaymentScore + needScore

Classification Matrix

The system uses a 2x2 decision matrix:

                HIGH NEED (≥300)         LOW NEED (<300)
                ┌────────────────┬──────────────────────┐
HIGH REPAYMENT  │  ✅ AUTO-APPROVE│  ⏳ MANUAL REVIEW    │
(≥350)          │                │                      │
                ├────────────────┼──────────────────────┤
LOW REPAYMENT   │  ⏳ MANUAL REVIEW│  ❌ REJECT          │
(<250)          │                │                      │
                └────────────────┴──────────────────────┘

Classification Rules:

  • ✅ Auto-Approve: High repayment (≥350) + High need (≥300)
  • ⏳ Manual Review: Medium scores or mixed high/low scores
  • ❌ Reject: Low repayment (<250) + Low need (<200)

For detailed explanation, see CLASSIFICATION_MATRIX.md

🧪 Testing

Test Cases

1. Auto-Approve Scenario ✅

Name: Test User 1
Phone: 9876543210
Aadhar: 1234 5678 9012
Category: Repeat Borrower
Loan Amount: ₹50,000
Previous Loans: 5
Electricity Bill: ₹500
Mobile Recharge: ₹100
Utility Payments: ₹300
Business Income: ₹30,000

Expected: Composite Score ~850, Auto-Approved

2. Reject Scenario ❌

Name: Test User 2
Phone: 8765432109
Aadhar: 9876 5432 1098
Category: New Borrower
Loan Amount: ₹100,000
Previous Loans: 0
Electricity Bill: ₹5,000
Mobile Recharge: ₹1,000
Utility Payments: ₹2,000
Business Income: ₹5,000

Expected: Composite Score ~300, Rejected

3. Manual Review Scenario ⏳

Name: Test User 3
Phone: 7654321098
Aadhar: 5432 1098 7654
Category: New Borrower
Loan Amount: ₹30,000
Previous Loans: 1
Electricity Bill: ₹1,200
Mobile Recharge: ₹300
Utility Payments: ₹800
Business Income: ₹15,000

Expected: Composite Score ~550, Manual Review

🔧 Configuration

Customizing Scoring Weights

Edit src/constants.js to adjust scoring parameters:

export const SCORING_WEIGHTS = {
  // Repayment Score Weights (0-500)
  BASE_REPAYMENT_SCORE: 150,
  PREVIOUS_LOAN_MULTIPLIER: 40,
  PREVIOUS_LOAN_MAX: 200,
  BUSINESS_INCOME_DIVISOR: 1000,
  BUSINESS_INCOME_MULTIPLIER: 5,
  BUSINESS_INCOME_MAX: 150,
  MAX_REPAYMENT_SCORE: 500,

  // Need Score Weights (0-500)
  ELECTRICITY_BASE: 200,
  ELECTRICITY_DIVISOR: 10,
  MOBILE_BASE: 150,
  MOBILE_DIVISOR: 5,
  UTILITY_BASE: 150,
  UTILITY_DIVISOR: 10,
  MAX_NEED_SCORE: 500
};

Loan Configuration

export const LOAN_CONFIG = {
  MIN_LOAN_AMOUNT: 1000,
  MAX_LOAN_AMOUNT: 500000,
  INTEREST_RATE: "4% APR",
  PROCESSING_TIME: "Same Day for Auto-Approved",
  REPAYMENT_PERIOD: "12-60 months"
};

🌐 API Endpoints

Applications

  • GET /api/applications - Fetch all applications
  • POST /api/applications - Create new application
  • GET /api/applications/:id - Get application by ID
  • PUT /api/applications/:id - Update application
  • DELETE /api/applications/:id - Delete application

Example Request

POST /api/applications
Content-Type: application/json

{
  "name": "Ramesh Kumar",
  "repaymentScore": 420,
  "needScore": 380,
  "compositeScore": 800,
  "status": "approved",
  "amount": 50000,
  "category": "Repeat Borrower",
  "classification": "High Need + Good Repayment"
}

🎨 UI Screenshots

Add screenshots of your application here:

  • Landing Page
  • Application Form
  • Score Result
  • Dashboard
  • Admin Panel

🔒 Security & Validation

Input Validation

  • Aadhar: 12-digit numeric validation
  • Phone: 10-digit mobile number (starts with 6-9)
  • Loan Amount: Between ₹1,000 and ₹5,00,000
  • File Upload: PDF, JPG, PNG only, max 5MB per file

Data Security

  • Client-side input sanitization
  • Server-side validation
  • localStorage backup for offline capability
  • API error handling and retry logic

♿ Accessibility

SahayScore is built with accessibility in mind:

  • ✅ ARIA labels on all interactive elements
  • ✅ Keyboard navigation support
  • ✅ Screen reader friendly
  • ✅ Focus indicators for keyboard users
  • ✅ Error announcements with role="alert"
  • ✅ Semantic HTML structure
  • ✅ Color contrast compliance (WCAG AA)

🚧 Roadmap

Phase 1 (Completed ✅)

  • Dual credit scoring system
  • 2x2 classification matrix
  • Application form with validation
  • Dashboard and admin panel
  • Backend API integration

Phase 2 (In Progress 🔄)

  • Database integration (MongoDB/PostgreSQL)
  • User authentication & authorization
  • Email/SMS notifications
  • PDF report generation
  • Advanced analytics dashboard

Phase 3 (Planned 📋)

  • Machine learning model training
  • Predictive default analysis
  • Mobile app (React Native)
  • Multi-language support
  • Blockchain-based credit history

🤝 Contributing

Contributions are welcome! Please follow these steps:

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

Development Guidelines

  • Follow existing code style and conventions
  • Add comments for complex logic
  • Update documentation for new features
  • Test thoroughly before submitting PR
  • Keep commits atomic and meaningful

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

👥 Team

Project: SahayScore - NBCFDC Loan Assessment Platform
Hackathon: Aviothic 2.0 - Debuggers Team
Organization: National Backward Classes Finance & Development Corporation

🙏 Acknowledgments

  • NBCFDC for the opportunity to innovate in financial inclusion
  • React and Vite communities for excellent tooling
  • Tailwind CSS for rapid UI development
  • Framer Motion for smooth animations
  • All open-source contributors

📞 Support

For questions, issues, or feedback:

Made with ❤️ for Financial Inclusion

⭐ Star this repo if you find it helpful! ⭐

About

ai-credit-scoring-system.vercel.app

Resources

Readme
Activity

Stars

0 stars

Watchers

0 watching

Forks

1 fork
Report repository

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •  

Languages