Custom Chatbot for any website with Retrieval Augmented Generation

This is the course project for the Applied Machine Learning Course of the M.Sc. Data Science program at the Chennai Mathematical Institute.

Group members:

• Roudranil Das
• Soham Sengupta
• Subhashree Saha
• Ujan Dasgupta

Summary

We scrape websites containing cooking recipes and extract the text content and store them. Then we build a RAG pipeline to get responses from LLM's based on queries, with context from these stored texts.

Detailed code for the rag pipeline is in the src directory.

Here is the link to the presentation.