Scribd
Upload
53 views4 pages

RAG AI Chatbot for Campus Documents

The document outlines a project to create a campus assistant chatbot that utilizes Retrieval-Augmented Generation (RAG) to answer questions from PDF documents such as course syllabi and student handbooks. It details the tools and technology stack required, including Python, PyMuPDF, and OpenAI GPT. The architecture involves loading PDFs, chunking them, generating embeddings, and using a language model to provide answers based on user queries.

Uploaded by

rendygunawan1604
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
53 views4 pages

RAG AI Chatbot for Campus Documents

The document outlines a project to create a campus assistant chatbot that utilizes Retrieval-Augmented Generation (RAG) to answer questions from PDF documents such as course syllabi and student handbooks. It details the tools and technology stack required, including Python, PyMuPDF, and OpenAI GPT. The architecture involves loading PDFs, chunking them, generating embeddings, and using a language model to provide answers based on user queries.

Uploaded by

rendygunawan1604
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd

CAMPUS ASSISTANT

WITH RAG
(RETRIEVAL-
AUGMENTED
GENERATION)
AI CHATBOT THAT ANSWERS QUESTIONS FROM PDFS

PRESENTED BY: RENDY


MY PROJECT IDEA

• Build a chatbot for campus documents


• PDFs: course syllabi, academic calendar,
student handbook
• Example: 'When does the semester start?'
• Example: 'What are the AI course topics?'
TOOLS AND STACK

• Python for coding


• PyMuPDF for reading PDFs
• sentence-transformers for embeddings
• Chroma for vector storage
• OpenAI GPT or HuggingFace for LLM
• LangChain or LlamaIndex for RAG
• Optional: Streamlit for chatbot UI
HOW IT WORKS
(ARCHITECTURE)
• 1. Load PDF documents
• 2. Split into chunks
• 3. Generate embeddings (sentence-
transformers)
• 4. Store in vector DB (Chroma)
• 5. User asks a question
• 6. Retrieve relevant chunks
• 7. LLM answers using retrieved text

You might also like