RAG-Based AI Chatbot for Navigating University Curriculum

To address the challenge students face in navigating scattered and extensive course information, this project presents a RAG-based AI-powered academic chatbot for University Curriculum. The system is built on a Retrieval-Augmented Generation (RAG) framework, grounding responses in official university documents.The process begins with creating a knowledge base by extracting data from online syllabi and PDFs using web scraping and text parsing. Documents are chunked and converted into semantic vectors via a pre-trained embedding model from Hugging Face, while a FAISS vector database enables fast and efficient retrieval through semantic search. When a student submits a query, the most relevant document excerpts are retrieved and passed to an open�source language model from Hugging Face, which synthesizes the context to generate clear, accurate, and natural language responses.The chatbot interface is built with Streamlit, providing a lightweight and self-contained
application that operates without external API dependencies. This system streamlines access to information on subjects, credit distribution, and curriculum details, demonstrating the integration of modern AI and NLP techniques to create a practical and impactful tool that enhances student self-reliance and academic support.