USDA Rural Opportunity Navigator 🚜

Overview

USDA Rural Opportunity Navigator is an AI-powered chatbot and analytics platform designed to connect rural Americans with crucial USDA Rural Development resources. Built in 24 hours for the HackMidwest-USDAChallenge, our solution leverages advanced web scraping, MongoDB vector search, and a Retrieval-Augmented Generation (RAG) model to deliver accurate, context-aware answers about USDA programs.

In addition to the chatbot, we provide a customizable, USDA-facing data analytics dashboard. This tool offers actionable insights from user interaction data, helping the agency better understand community needs and resource engagement.

---

Features

• Conversational AI Chatbot: Answers questions about USDA Rural Development programs using RAG and OpenAI models.
• Web Scraping Pipeline: Extracts and updates USDA.gov program data for the knowledge base.
• MongoDB Vector Search: Enables semantic retrieval of relevant program info.
• Data Analytics Dashboard: Visualizes user engagement, satisfaction, and trending topics for USDA analysts.
• Location-Aware Matching: Suggests state-specific resources based on user location.

---

Architecture

┌────────────┐     ┌──────────────┐     ┌──────────────┐     ┌──────────────┐
│  Frontend  │ <-> │ FastAPI API  │ <-> │ MongoDB Atlas│ <-> │ Web Scraper  │
│ (React)    │     │ (Python)     │     │ (Vector DB)  │     │ (Python)     │
└────────────┘     └──────────────┘     └──────────────┘     └──────────────┘
        │
        ▼
┌────────────────────────────┐
│ Data Analytics Dashboard   │
│ (React + Plotly)          │
└────────────────────────────┘

---

Quickstart

1. Clone the Repo

git clone https://github.com/Ape108/USDAChatBot.git
cd USDAChatBot

2. Backend Setup (Python)

a. Create & Activate Virtual Environment

./scripts/setup_venv.sh
source .venv/bin/activate

b. Install Dependencies

pip install -r backend/requirements.txt

c. Set Environment Variables

Create a .env file in backend/ with:

OPENAI_API_KEY=your-openai-key
MONGO_URI=your-mongodb-uri

d. Run the API Server

cd backend
uvicorn main_backend:app --reload

3. Frontend Setup (React + Vite)

cd frontend
npm install
npm run dev

The app will be available at http://localhost:5173 by default.

---

Components

Chatbot (AI + RAG)

• Backend: backend/main_backend.py, backend/usda_chat_api.py, backend/AItester/usda_vector_search.py
• Frontend: frontend/src/components/Chatbot/Chatbot.jsx
• Vector Search: Uses OpenAI embeddings + MongoDB Atlas for semantic retrieval.

Data Analytics Dashboard

• Backend: backend/data-analytics/
  • Loads and processes chatbot_interaction_data.csv
  • Generates interactive Plotly visualizations (KPIs, trends, maps, satisfaction, etc.)
• Frontend: frontend/src/components/Dashboard/Dashboard.jsx

Web Scraper

• Scripts: backend/webscraper/
  • Scrapes USDA.gov sitemaps and program pages
  • Populates MongoDB with up-to-date program info and embeddings

---

Data & Documentation

• Data Dictionary: backend/data-analytics/docs/data_dictionary.md
• Data Sources: backend/data-analytics/docs/data_sources.md
• Data Contract: backend/data-analytics/docs/data_contract.md

---

Example User Data Schema

See backend/data-analytics/chatbot_interaction_data.csv for sample fields:

session_id	user_id	timestamp	user_location_state	user_query	bot_response_summary	user_feedback	topic_category
SESS001	USER112	2025-10-01T08:05:30Z	TX	I need a loan...	Provided an overview...	1	Economic Development

---

Contributing

Pull requests and feedback are welcome! For major changes, please open an issue first to discuss what you would like to change.

---

License

This project is for demonstration purposes at CrackMidwest 2025. See individual files for third-party licenses.