GitHub - Fatheen/live-rag-chatbot-project: Live Website Retrieval (Live RAG), Dual Mode Support (CSV / Live)

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 8 Commits
.cache		.cache
.ipynb_checkpoints		.ipynb_checkpoints
.ipython/profile_default		.ipython/profile_default
.jupyter		.jupyter
.keras		.keras
.local/share/jupyter		.local/share/jupyter
.npm/_logs		.npm/_logs
.streamlit		.streamlit
__pycache__		__pycache__
.gitconfig		.gitconfig
.gitignore		.gitignore
ACRONYM_EXAMPLES.md		ACRONYM_EXAMPLES.md
ACRONYM_EXTRACTION_ANALYSIS.md		ACRONYM_EXTRACTION_ANALYSIS.md
ACRONYM_QUERY_ANALYSIS.md		ACRONYM_QUERY_ANALYSIS.md
RAG testing questions.docx		RAG testing questions.docx
Webscraping.ipynb		Webscraping.ipynb
api.py		api.py
chatbot_rag.py		chatbot_rag.py
emb_cache.npz		emb_cache.npz
live_rag.py		live_rag.py
page_cache.json		page_cache.json
readme.txt		readme.txt
requirements.txt		requirements.txt
streamlit_app.py		streamlit_app.py
uiuc_complete_content_test.csv		uiuc_complete_content_test.csv

Repository files navigation

# RAG Chatbot for Business & Finance Website

This project is a chatbot system that uses **Retrieval-Augmented Generation (RAG)** to answer questions based on content from the University of Illinois Business & Finance website.

It includes:
- A **FastAPI** backend (`api.py`)
- A **retrieval + generation module** (`chatbot_rag.py`)
- A **Streamlit UI** (`streamlit_app.py`)
- A **CSV file** with scraped and cleaned website content


HOW TO RUN:
Run the Backend (FastAPI)
uvicorn api:app --reload --host 0.0.0.0 --port 8000

Run the Frontend (Streamlit)
streamlit run streamlit_app.py

Access the UI
https://jupyter.gpu.atlas.illinois.edu/user/<NetID@illinois.edu>/proxy/8501/?redirects=1

50 test questions:
https://docs.google.com/document/d/1i_jmUYqEdjwQX0v2oM96r2h2Dny4aZDDqqzz7vBJVDA/edit?usp=sharin