MultiPDF Chatbot

Overview

MultiPDF Chatbot is a Streamlit application that allows users to interact with multiple PDF documents through a conversational interface. The application leverages various libraries and tools to process PDFs, extract text, and enable users to ask questions about the content of their documents.

Features

• PDF Upload: Users can upload multiple PDF documents.
• Text Extraction: The application extracts text from the uploaded PDFs.
• Conversational Interface: Users can ask questions about the content of their PDFs and receive answers in a chat-like interface.
• Vector Store: The application uses a vector store to manage and retrieve text chunks efficiently.
• Memory Management: The conversation history is managed to provide context-aware responses.

How It Works

1. PDF Upload: Users upload their PDF documents through the Streamlit interface.
2. Text Extraction: The text from the PDFs is extracted using the PyPDF2 library.
3. Text Chunking: The extracted text is split into manageable chunks using the CharacterTextSplitter from the langchain library.
4. Vector Store: The text chunks are embedded using GoogleGenerativeAIEmbeddings and stored in a vector store using FAISS.
5. Conversational Chain: A conversational retrieval chain is created using the ConversationalRetrievalChain from the langchain library, which retrieves relevant text chunks based on the user's questions.
6. User Interaction: Users can ask questions, and the application provides answers based on the retrieved text chunks.

Dependencies

The application relies on the following dependencies:

• streamlit: For creating the web interface.
• PyPDF2: For extracting text from PDFs.
• langchain: For text splitting, embedding, and conversational retrieval.
• faiss: For efficient similarity search and clustering.
• dotenv: For loading environment variables.

Usage

1. Clone the repository:

   git clone <repository_url>

2. Install the required dependencies:

   pip install -r requirements.txt

3. Run the application:

   streamlit run app.py

4. Upload your PDF documents and start asking questions.