AgriHub

AgriHub is an AI-powered agricultural marketplace and credit platform designed to help farmers:

• Sell produce more efficiently
• Build verifiable transaction histories
• Access loans based on real economic activity

The platform combines:

• Marketplace (farmers ↔ buyers)
• Machine Learning (credit scoring)
• Payment integration (loan disbursement & repayment)

---

Problem

Millions of farmers:

• Sell products informally and struggle to find competitive buyers
• Lack financial records
• Cannot access loans

Traditional lenders require:

• Bank history
• Credit score
• Collateral

Most farmers do not have these.

---

Solution

AgriHub creates a digital financial identity for farmers using:

• Transaction activity
• Marketplace behavior
• Repayment patterns

It then uses machine learning to:

• Predict loan eligibility
• Recommend safe loan amounts

---

Key Idea

"Your farming activity becomes your credit score"

---

System Architecture

Frontend (Next.js)
↓
Backend Services
├── ML Credit Scoring API (FastAPI)
├── Payment Integration (Interswitch)
↓
Cloud Services (Firebase)

---

Core Components

1. Marketplace

Farmers can:

• Create listings
• Upload produce images
• Set prices and quantities

Buyers can:

• Browse products
• View details
• Place orders

---

2. Orders & Transactions

The system tracks:

• Completed sales
• Transaction values
• Frequency of activity

This data is used for credit scoring.

---

3. Payments

Planned flow:

• Buyers pay for produce
• Farmers receive funds
• Loan repayment processing

Supports:

• Direct payments
• integration with providers (e.g., Interswitch)

---

4. ML Credit Scoring (Agrihub-ML)

Separate repository:
https://github.com/Jaykay73/Agrihub-ML

This service:

• Receives farmer activity data
• Processes it using a trained model
• Returns:

{
  "credit_score": 78,
  "loan_eligible": true,
  "probability_of_eligibility": 0.78
}

Features Used by the Model

• Total sales
• Number of transactions
• Average transaction value
• Activity frequency
• Repayment history
• High-value transactions
• Blockchain verification ratio

Output

• Credit score (0–100)
• Loan eligibility
• Suggested loan amount
• Risk level

---

6. Loan System

Flow

1. Farmer builds transaction history
2. ML model evaluates risk
3. Loan is approved or rejected

If approved:

• Funds are disbursed via payment API
• Farmer repays over time
• Repayment updates future scores

---

User Roles

Farmer

• Lists produce
• Receives orders
• Builds credit history
• Applies for loans

Buyer

• Browses marketplace
• Purchases produce
• Interacts with sellers

Platform / Lender

• Evaluates credit scores
• Approves loans
• Manages repayments

---

Tech Stack

Frontend

• Next.js
• React
• TypeScript
• Tailwind CSS
• Framer Motion

Backend / Services

• FastAPI (ML API)
• Python (ML models)
• Firebase (data storage)

Machine Learning

• scikit-learn
• pandas
• numpy

Other Tools

• React Query (data fetching)

---

Project Structure

AgriHub/
├── app/                # Next.js routes (buyer, farmer, auth)
├── features/           # Business logic modules
│   ├── marketplace/
│   ├── loans/
│   ├── payments/
│   ├── orders/
│   └── listings/
├── lib/                # Utilities (firebase, cloudinary, etc.)
├── shared/             # Shared components & types
├── ml-service/         # Local ML service copy

---

Running the Project

Frontend

npm install
npm run dev

Open: http://localhost:3000

---

ML Backend

cd agrihub-ml

pip install -r requirements.txt
uvicorn app:app --reload

Open: http://127.0.0.1:8000/docs

---

API Integration

Example:

fetch("/credit-score", {
  method: "POST",
  body: JSON.stringify(data)
});

Impact

AgriHub aims to:

• Empower farmers financially
• Reduce loan risk using data
• Digitize agricultural trade
• Create fair access to credit

---

Key Innovation

AgriHub does not just sell farm produce.

It turns economic activity into financial identity.

---

Author

Built by: Aledare John Oluwaseun (ML Engineer) Onoja Possible (Fullstack Engineer)