1. Introduction
This AI Chat application is designed to assist users by answering their questions using Retrieval-Augmented Generation (RAG) technology. It leverages a knowledge base from files, ensuring responses are both accurate and contextually relevant. Organizations can integrate this chatbot to provide instant and reliable answers to employees, customers, or stakeholders. By retrieving information from structured documents, manuals, and FAQs, it enhances efficiency and accessibility. The system can handle diverse queries, from company policies to technical documentation. Unlike traditional chatbots, this AI continuously learns and improves based on its knowledge base. Additionally, it can remember user questions based on chat history, allowing for a more personalized and contextual conversation. This ensures a smoother user experience, as the chatbot can reference past interactions to provide more accurate and relevant answers.

2.Dataset Description
The dataset used in this project consists of the following:
Sample questions and answers in text file.
The dataset is stored in local vector store with documents pre-processed

3.Experiments
The following are the steps performed to build the Chatbot.

1. Document Pre-Processing
   Documents were preprocessed using Langchain’s text splitter to create chunks of optimal size for retrieval.
2. RAG implementation with Open AI LLM

• Contextualizing question based on chat history
  
• Create prompt with contextualized question
  
• Chain LLM and new created prompt
  
• Invoke chain
  

4.Result