National Carbon Credit Fraud Detection Platform

Solving the Trust Crisis in Carbon Markets through AI & Transparency.

🌍 The Problem

The voluntary carbon market is plagued by double-counting, inflated claims, and phantom projects.

• Companies claim credits for forests that don't exist.
• Multiple entities sell the same credit twice.
• Auditors lack real-time data to verify claims.

🌐 Live Demo

Try the platform here: https://qmicservices.vercel.app

💡 The Solution

This platform introduces a Digital Oversight Layer that sits between Companies and the Government/Public. It solves fraud through three key mechanisms:

1. Trusted Verification Workflow (RBAC)

We replaced the "honesty system" with a strict Government-Gatekeeper Model.

• Entities cannot act until verified by a Government Authority.
• Role Separation:
  • Company: Submits raw data.
  • Government: Validates legal standing.
  • AI Engine: Validates physical reality.

2. AI-Driven "Reality Check" (Simulation)

Every credit submission undergoes an automated Cross-Reference Analysis:

• Input: User claims 5000 tons of CO2e from "Amazon Sector 4".
• Logic: The system (simulated) pulls satellite vegetation indices (NDVI) for that coordinate.
• Output: A Risk Score (0-100).
  • High Risk: Discrepancy between claim and satellite data. -> Flagged for Investigation.
  • Low Risk: Consistent data. -> Auto-Approved.

3. Public Transparency

Sunlight is the best disinfectant. The Public Registry allows any citizen or auditor to:

• View "Qualified" vs "Pending" projects.
• See the audit trail of who approved a project and when.
• File anonymous complaints if they suspect local malpractice.

---

🚀 Quick Start Guide

Prerequisites

1. Node.js (v18+)
2. MongoDB (Local or Atlas)

Installation & Run

The project is split into backend (API/Data) and frontend (UI/Dashboards).

1. Start Backend (The Brain)

cd backend
npm install
npm run seed     # ⚠️ IMPORTANT: Seeds realistic "Fraud" & "Verified" demo scenarios
npm run dev

2. Start Frontend (The Interface)

cd frontend
npm install
npm run dev

Access the app at: http://localhost:5173

---

🔑 Demo Scenarios (Problem-Solving Showcases)

Log in to experience the different sides of the solution:

🕵️‍♂️ Scenario A: The Government Overser

User: government@user.in | Password12

• Goal: Detect Anomalies.
• Action:
  1. Go to Dashboard.
  2. Check the "Live Monitor" tab.
  3. See real-time "High Severity" flags generated by the seed script.
  4. Use the Filter button to focus on "High Priority" risks.

🏭 Scenario B: The Compliant Company

User: company@user.in | Password12

• Goal: Submit a Legitimate Claim.
• Action:
  1. Go to New Submission.
  2. Upload a "Evidence PDF" (Simulation).
  3. Watch the AI Analysis Toast verify the document.
  4. See the immediate Risk Score feedback.

👁️ Scenario C: The Public Auditor

User: people@user.in | Password

• Goal: Verify Transparency.
• Action:
  1. Go to Public Registry.
  2. Search for verified projects.
  3. Confirm that "High Risk" projects are NOT listed publicly.

---

🛠️ Architecture & Tech Stack

While the focus is on the solution, the technology enables the scale:

• Frontend: React (Vite) + Tailwind CSS (for rapid, professional UI).
• UI/UX: Default Light Mode for clean visibility, with Recharts for data visualization.
• Backend: Node.js + Express + MongoDB (for flexible schema evolution).
• Security: JWT Auth + Role-Based Guards (Middleware).
• Communication: REST APIs with Axios Interceptors for reliable data fetching.
• Quality: TypeScript everywhere to prevent logic errors.

---

📂 Project Structure

• backend/src/scripts/seed.ts: The "Simulation Engine" that generates the fraud scenarios.
• frontend/src/features/gov: The Oversight Dashboards.
• frontend/src/features/company: The Compliance & Submission tools.