🇮🇳 PM-AJAY Digital Platform (SIHPS25152)

A comprehensive digital platform for the Pradhan Mantri Anusuchit Jaati Abhyudaya Yojana (PM-AJAY) scheme, built for Smart India Hackathon 2025. This platform modernizes welfare fund disbursement, beneficiary management, and evidence tracking using blockchain technology, AI/ML models, and multi-channel communication.

🏗️ Architecture Overview

┌─────────────────────────────────────────────────────────────────────────────────┐
│                           PM-AJAY Digital Platform                              │
├─────────────┬─────────────┬──────────────┬──────────────┬──────────────────────┤
│   Frontend  │   Backend   │  Blockchain  │   AI Models  │   Communication      │
│  (Next.js)  │ (Node/Fast) │  (Fabric)    │   (Python)   │  (WhatsApp/Chatbot)  │
│  Port:3000  │ Port:1604   │  Port:3002   │  Port:5000+  │     Port:3000        │
└─────────────┴─────────────┴──────────────┴──────────────┴──────────────────────┘
         │            │              │              │               │
         └────────────┴──────────────┴──────────────┴───────────────┘
                                     │
                              ┌──────┴──────┐
                              │   MongoDB   │
                              │  Port:27017 │
                              └─────────────┘

🛠️ Tech Stack

Component	Technologies
Frontend	Next.js 14, React 18, TailwindCSS, TypeScript, Shepherd.js (guided tours)
Backend (Main API)	Node.js 18+, Express.js, MongoDB, Mongoose, JWT Auth
Backend (Translation)	FastAPI, Python 3.9+, Google Translate API (deep-translator)
Blockchain	Hyperledger Fabric 2.x, Spring Boot (Java 17), Docker
AI/ML Models	Python, XGBoost, Scikit-learn, LangChain, ChromaDB, RAG
Chatbot	Node.js, Twilio WhatsApp API, MongoDB, Tesseract OCR
Video Conferencing	Zego Cloud SDK

✨ Features

🖥️ Frontend Dashboard

Feature	Description
User Authentication	Secure login/signup with role-based access (Ministry, PACC, State, SNA, IA, SAU)
Dashboard Analytics	Real-time KPIs with donut charts, time-series graphs, and India heatmap
Proposal Management	Create, track, and approve welfare proposals with multi-stage workflow
Project Tracking	Monitor milestones with visual tracker and evidence uploads
Funds Management	Track FTOs, PFMS confirmations, bank acknowledgments, and reconciliation
Beneficiary Management	View/filter beneficiaries with ranking scores and demographic data
Evidence Repository	Upload photos/documents with blockchain hash verification
Reports Generation	Generate PDF/Excel reports for audits and compliance
SNA Dashboard	State Nodal Agency view with account management
PACC View	Project Appraisal & Central Committee approval interface
Multi-language Support	14+ Indian languages (Hindi, Bengali, Tamil, Telugu, etc.)
Video Conferencing	Built-in Zego meeting rooms for virtual discussions
AI Chatbot	Scheme FAQ assistant with context-aware responses
Guided Tour	Interactive onboarding with Shepherd.js

⚙️ Backend APIs

Main Backend (Node.js/Express) - Port 1604

User Management (Register, Login, Profile, OAuth)
Proposal CRUD with approval workflow
Project lifecycle with milestone tracking
Utilization Certificate (UC) upload/verification
Central ministry data aggregation
Document hash verification

Translation API (FastAPI) - Port 8000

Real-time translation for 14+ Indian regional languages
Auto-detect source language
Google Translate integration via deep-translator

⛓️ Blockchain (Hyperledger Fabric) - Port 3002

Channel	Purpose	Organizations
Approval	Proposal & blueprint hashes	Ministry, PACC, State
Funds	FTO, PFMS, bank records	Ministry, State, SNA
Implementation	Beneficiary & evidence hashes	State, IA, SNA
Audit	Audit reports & UC hashes	SAU, State, Ministry

20 blockchain API endpoints for hash storage and verification across 4 channels.

🤖 AI/ML Models

Model	Purpose	Data Files
Beneficiary Ranking	XGBoost priority scoring	`synthetic_beneficiaries.csv` (1.7MB), `trained_model.json`
Question Clustering	Grievance pattern analysis	`pm_ajay_final_simulated_data.csv`, `processed_issues_uuid.csv`
State Plan RAG	Document Q&A with LangChain	`input.json`, `input2.json`

📱 WhatsApp Digital Khata Bot

User registration with demographic collection
Digital ledger for credits/debits
Transaction categories: Income, Expense, Loans, Repayments, Savings
OCR support for document scanning (Tesseract)
Twilio WhatsApp integration

📁 Project Structure

SIHPS25152_FINAL/
│
├── frontend/                         # Next.js 14 Frontend (Port 3000)
│   ├── app/                          # App Router pages
│   │   ├── dashboard/                # Main dashboard with charts
│   │   ├── beneficiary/              # Beneficiary management
│   │   ├── funds/                    # FTO and fund tracking
│   │   ├── projects/                 # Project milestone tracking
│   │   ├── proposals/                # Proposal workflow
│   │   ├── reports/                  # Report generation
│   │   ├── sna/                      # State Nodal Agency view
│   │   ├── pacc/                     # PACC approval interface
│   │   ├── login/                    # Authentication
│   │   └── signup/                   # User registration
│   ├── components/                   # 36 reusable UI components
│   │   ├── chatbot.tsx               # AI assistant (28KB)
│   │   ├── sidebar.tsx               # Navigation (23KB)
│   │   ├── india-map-heatmap.tsx     # State visualization
│   │   └── ...
│   ├── lib/                          # Utilities and helpers
│   └── public/                       # Static assets
│
├── backend/                          # Main Backend (Port 1604)
│   ├── src/
│   │   ├── controllers/              # 7 request handlers
│   │   ├── models/                   # 10 MongoDB schemas
│   │   │   ├── beneficiary.model.js
│   │   │   ├── proposal.model.js
│   │   │   ├── project.model.js
│   │   │   ├── ngo.model.js
│   │   │   └── ...
│   │   ├── routes/                   # 6 API route files
│   │   └── middlewares/              # Auth & validation
│   ├── main.py                       # FastAPI Translation Server (Port 8000)
│   └── requirements.txt              # Python dependencies
│
├── app-backend/                      # Secondary MERN Backend
│   └── src/                          # Controllers, models, routes
│
├── blockchain/                       # Hyperledger Fabric (Port 3002)
│   ├── backend/                      # Node.js blockchain gateway
│   ├── backend-spring/               # Spring Boot API (Java 17)
│   ├── blockchain/                   # Chaincode & network config
│   │   ├── organizations/            # Org certificates
│   │   └── docker/                   # Docker compose files
│   ├── API_REFERENCE.md              # Complete API docs (20 endpoints)
│   └── commands.sh                   # Setup scripts
│
├── AI_Model/                         # Machine Learning Models
│   ├── Individual Beneficiary Ranking Model/
│   │   ├── 1_synthesize_data.py      # Generate training data
│   │   ├── 2_train_model.py          # Train XGBoost model
│   │   ├── 3_test_model.py           # Model testing
│   │   ├── 6_app.py                  # Flask API (Port 5000)
│   │   ├── synthetic_beneficiaries.csv   # 1.7MB training data
│   │   └── trained_model.json        # Trained XGBoost model
│   │
│   ├── Question Clustering/
│   │   ├── app.py                    # Clustering API
│   │   ├── clusters.py               # K-means implementation
│   │   └── pm_ajay_final_simulated_data.csv
│   │
│   └── State Plan Analysis RAG/
│       ├── main.py                   # RAG entry point
│       ├── rag_system.py             # LangChain RAG implementation
│       ├── analysis agent.py         # Document analysis
│       └── tools.py                  # Utility functions
│
├── WHATSAPP/                         # WhatsApp Khata Bot (Port 3000)
│   ├── src/
│   │   ├── controllers/              # Message handlers
│   │   ├── models/                   # User & Khata schemas
│   │   └── services/                 # Business logic
│   ├── config/                       # Twilio configuration
│   ├── eng.traineddata               # English OCR (5.2MB)
│   └── hin.traineddata               # Hindi OCR (1.6MB)
│
├── pmajay-chatbot/                   # Additional Chatbot Service
│   └── chatbot_backend/
│       └── chroma_db/                # Vector database for RAG
│
├── .gitignore                        # Excludes venv, node_modules, large files
└── README.md                         # This file

📊 Data Files

AI Model Training Data

File	Location	Size	Description
`synthetic_beneficiaries.csv`	`AI_Model/Individual Beneficiary Ranking Model/`	1.7 MB	10,000+ synthetic beneficiary records for training
`location dataset.csv`	`AI_Model/Individual Beneficiary Ranking Model/`	4 KB	State/district location mappings
`trained_model.json`	`AI_Model/Individual Beneficiary Ranking Model/`	620 KB	Trained XGBoost model weights
`preprocessor.joblib`	`AI_Model/Individual Beneficiary Ranking Model/`	4 KB	Sklearn preprocessing pipeline
`pm_ajay_final_simulated_data.csv`	`AI_Model/Question Clustering/`	100 KB	Simulated grievance/issue data
`processed_issues_uuid.csv`	`AI_Model/Question Clustering/`	45 KB	Processed issues with UUIDs
`input.json`	`AI_Model/State Plan Analysis RAG/`	3.6 KB	Sample state plan input
`input2.json`	`AI_Model/State Plan Analysis RAG/`	5 KB	Additional plan data

OCR Training Data (WhatsApp Bot)

File	Size	Description
`eng.traineddata`	5.2 MB	Tesseract English OCR model
`hin.traineddata`	1.6 MB	Tesseract Hindi OCR model

🚀 Getting Started

Prerequisites

Tool	Version	Purpose
Node.js	v18+	Backend & Frontend
Python	3.9+	AI Models & Translation API
MongoDB	6.0+	Database
Docker	24.0+	Blockchain network
Java	17+	Spring Boot blockchain API
pnpm or npm	Latest	Package management

Quick Install Check

node --version    # Should be v18+
python --version  # Should be 3.9+
mongod --version  # Should be 6.0+
docker --version  # Should be 24.0+
java --version    # Should be 17+

📝 Step-by-Step Run Instructions

Method 1: Run Core Services Only (Recommended for Quick Start)

Open 4 terminal windows and run each command in sequence:

Terminal 1: Start MongoDB

# Windows
mongod

# macOS/Linux
sudo mongod --dbpath /data/db

Terminal 2: Start Main Backend (Port 1604)

cd backend
npm install
npm run dev

✅ Expected output: Server is running on http://localhost:1604

Terminal 3: Start Translation API (Port 8000)

cd backend
pip install -r requirements.txt
python main.py

✅ Expected output: Uvicorn running on http://0.0.0.0:8000

Terminal 4: Start Frontend (Port 3000)

cd frontend
npm install    # or pnpm install
npm run dev    # or pnpm dev

✅ Expected output: Ready on http://localhost:3000

Access the Application

Open your browser and go to: http://localhost:3000

Method 2: Run All Services (Full Platform)

Step 1: Start MongoDB

mongod

Step 2: Start Main Backend

cd backend
npm install
npm run dev
# Runs on http://localhost:1604

Step 3: Start Translation API

cd backend
pip install -r requirements.txt
python main.py
# Runs on http://localhost:8000

Step 4: Start Frontend

cd frontend
npm install
npm run dev
# Runs on http://localhost:3000

Step 5: Start Blockchain Backend (Choose One)

Option A: Spring Boot (Recommended)

cd blockchain/backend-spring
./mvnw spring-boot:run    # Linux/macOS
mvnw.cmd spring-boot:run  # Windows
# Runs on http://localhost:3002

Option B: Node.js

cd blockchain/backend
npm install
npm start
# Runs on http://localhost:3002

Step 6: Start WhatsApp Bot (Optional)

cd WHATSAPP
npm install
npm run dev
# Runs on http://localhost:3000 (use different port if frontend is running)

⚠️ Note: For WhatsApp bot to work, you need ngrok and Twilio credentials configured.

Step 7: Run AI Models (Optional)

Beneficiary Ranking API (Port 5000):

cd "AI_Model/Individual Beneficiary Ranking Model"
pip install flask pandas xgboost scikit-learn joblib
python 6_app.py

Question Clustering:

cd "AI_Model/Question Clustering"
pip install flask pandas scikit-learn
python app.py

State Plan RAG:

cd "AI_Model/State Plan Analysis RAG"
pip install langchain chromadb openai
python main.py

Method 3: Using Docker (Coming Soon)

# docker-compose.yml will be added for one-command startup
docker-compose up -d

🔐 Environment Variables

Create .env files in each directory:

`frontend/.env`

NEXT_PUBLIC_API_URL=http://localhost:1604
NEXT_PUBLIC_TRANSLATION_API=http://localhost:8000
NEXT_PUBLIC_BLOCKCHAIN_API=http://localhost:3002

`backend/.env`

PORT=1604
MONGODB_URI=mongodb://localhost:27017/pmajay
CORS_ORIGIN=http://localhost:3000
JWT_SECRET=your_super_secret_jwt_key_here

`WHATSAPP/.env`

TWILIO_ACCOUNT_SID=your_twilio_sid
TWILIO_AUTH_TOKEN=your_twilio_token
TWILIO_WHATSAPP_NUMBER=whatsapp:+14155238886
MONGODB_URI=mongodb://localhost:27017/whatsapp_khata
PORT=3001

`blockchain/backend-spring/application.properties`

server.port=3002

📚 API Documentation

Quick API Reference

Service	Port	Base URL	Health Check
Frontend	3000	http://localhost:3000	Browser
Main Backend	1604	http://localhost:1604/api	`/api/health`
Translation API	8000	http://localhost:8000	`/health`
Blockchain API	3002	http://localhost:3002	`/health`

Main Backend Endpoints

POST   /api/users/register     - Register new user
POST   /api/users/login        - User authentication
GET    /api/proposals          - Get all proposals
POST   /api/proposals          - Create new proposal
GET    /api/projects           - Get all projects
POST   /api/projects           - Create new project
GET    /api/central            - Get central dashboard data
POST   /api/uc/upload          - Upload utilization certificate
GET    /api/health             - Health check

Translation API Endpoints

POST   /translate              - Translate text between languages
GET    /languages              - Get supported languages list
GET    /health                 - Health check

Blockchain API Endpoints

See blockchain/API_REFERENCE.md for complete documentation of 20 endpoints.

🧪 Quick Verification

After starting all services, run these commands to verify:

# Check main backend
curl http://localhost:1604/api/health
# Expected: {"status":"ok","message":"Backend is running","port":"1604"}

# Check translation API
curl http://localhost:8000/health
# Expected: {"status":"healthy","provider":"Google Translate"}

# Check blockchain API
curl http://localhost:3002/health
# Expected: {"status":"UP"}

# Test translation
curl -X POST http://localhost:8000/translate \
  -H "Content-Type: application/json" \
  -d '{"text":"Hello, how are you?","source_lang":"en","target_lang":"hi"}'

🐛 Troubleshooting

Common Issues

Issue	Solution
`EADDRINUSE: address already in use`	Kill process on that port: `npx kill-port 3000`
`MongoDB connection failed`	Ensure MongoDB is running: `mongod`
`Module not found`	Run `npm install` or `pip install -r requirements.txt`
`CORS error`	Check `CORS_ORIGIN` in backend `.env` matches frontend URL
`Port 3000 conflict (WhatsApp vs Frontend)`	Change WhatsApp port to 3001 in its `.env`

Windows-Specific

# Kill process on port
netstat -ano | findstr :3000
taskkill /PID <PID> /F

# Run Python scripts
python main.py  # Not python3

macOS/Linux-Specific

# Kill process on port
lsof -ti:3000 | xargs kill -9

# Run Python scripts
python3 main.py

👥 Access Roles

Role	Permissions
Ministry	Full access, fund release approval, audit oversight
PACC	Proposal approval, blueprint review
State	State-level project implementation, UC submission
SNA	Fund tracking, bank account management
IA (Implementing Agency)	Evidence upload, milestone updates
SAU (State Audit Unit)	Audit reports, corrective actions
Beneficiary	View benefits, raise issues via WhatsApp

📄 License

ISC License

🙏 Acknowledgments

Built for Smart India Hackathon 2025 to digitize and streamline the PM-AJAY scheme for efficient welfare fund distribution to Scheduled Caste beneficiaries.

Problem Statement ID: SIH PS 25152

🔗 Links

GitHub: https://github.com/codewithabhay10/SIH-Project-Sahay
Smart India Hackathon: https://www.sih.gov.in/

Made with ❤️ for Digital India 🇮🇳

Name		Name	Last commit message	Last commit date
Latest commit History 17 Commits
AI_Model		AI_Model
WHATSAPP		WHATSAPP
app-backend		app-backend
backend		backend
frontend		frontend
my-app		my-app
pmajay-chatbot		pmajay-chatbot
.gitattributes		.gitattributes
.gitignore		.gitignore
README.md		README.md

Folders and files

Latest commit

History

Repository files navigation

🇮🇳 PM-AJAY Digital Platform (SIHPS25152)

📋 Table of Contents

🏗️ Architecture Overview

🛠️ Tech Stack

✨ Features

🖥️ Frontend Dashboard

⚙️ Backend APIs

Main Backend (Node.js/Express) - Port 1604

Translation API (FastAPI) - Port 8000

⛓️ Blockchain (Hyperledger Fabric) - Port 3002

🤖 AI/ML Models

📱 WhatsApp Digital Khata Bot

📁 Project Structure

📊 Data Files

AI Model Training Data

OCR Training Data (WhatsApp Bot)

🚀 Getting Started

Prerequisites

Quick Install Check

📝 Step-by-Step Run Instructions

Method 1: Run Core Services Only (Recommended for Quick Start)

Terminal 1: Start MongoDB

Terminal 2: Start Main Backend (Port 1604)

Terminal 3: Start Translation API (Port 8000)

Terminal 4: Start Frontend (Port 3000)

Access the Application

Method 2: Run All Services (Full Platform)

Step 1: Start MongoDB

Step 2: Start Main Backend

Step 3: Start Translation API

Step 4: Start Frontend

Step 5: Start Blockchain Backend (Choose One)

Step 6: Start WhatsApp Bot (Optional)

Step 7: Run AI Models (Optional)

Method 3: Using Docker (Coming Soon)

🔐 Environment Variables

frontend/.env

backend/.env

WHATSAPP/.env

blockchain/backend-spring/application.properties

📚 API Documentation

Quick API Reference

Main Backend Endpoints

Translation API Endpoints

Blockchain API Endpoints

🧪 Quick Verification

🐛 Troubleshooting

Common Issues

Windows-Specific

macOS/Linux-Specific

👥 Access Roles

📄 License

🙏 Acknowledgments

🔗 Links

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

`frontend/.env`

`backend/.env`

`WHATSAPP/.env`

`blockchain/backend-spring/application.properties`

Packages