RAG-based-Intelligent-Conversational-AI-Agent-for-Knowledge-Extraction-Using-LangChain-Gemini-LLM

### RAG based Intelligent Conversational AI Agent for Knowledge Extraction using Langchain Gemini LLM


<div align ="center">

[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1rUJ_wBEYFZsFijDzjOjI8QD9IBeLfB2s?usp=sharing)

</div>
<div align ="center">

### In the above google colab contain detailed code 
</div>


<div align ="center">

   
![image](https://miro.medium.com/v2/resize:fit:1400/format:webp/1*SuS0D_-uSf5RnLuUlqgytw.png)

</div>



Retrieval-Augmented Generation (RAG) is a framework that combines information retrieval with generative AI. It allows models to retrieve relevant information from external sources or databases and use that data to generate more accurate and contextually relevant responses. By leveraging both retrieval and generation, RAG improves the accuracy and reliability of AI models, particularly in providing up-to-date information or handling complex questions.

## **Workflow**

This project provides an AI-based conversational assistant that leverages Retrieval-Augmented Generation (RAG) to extract knowledge from PDF documents. The system combines text embeddings, vector search, and an LLM to provide answers to user questions. Below is a detailed step-by-step workflow of how the application operates:

### 1. **Uploading the PDF Document**
   - Users upload a PDF file through the path mentioning on notebook. The uploaded file is processed to extract the text using `pdfplumber`, a Python library for extracting text from PDFs.
   
### 2. **Text Extraction**
   - The Notebook utilizes the `pdfplumber` library to extract raw text from the uploaded PDF. Each page of the document is parsed, and the resulting text is prepared for further processing.

### 3. **Text Chunking**
   - The extracted text is split into smaller chunks using `RecursiveCharacterTextSplitter`. This ensures the content is manageable for embeddings and retrieval, typically with a chunk size of 500 characters