Chatbot with RAG (Retrieval-Augmented Generation)

chatbot-documents.mp4

📝 Description

This project is a Chatbot with RAG (Retrieval-Augmented Generation) that uses the FAISS RAG method and LangChain to provide intelligent and context-aware responses. The chatbot is designed to:

• Understand your queries in Portuguese.
• Read and process PDF documents uploaded by users.
• Allow users to interact with their documents using AI-powered conversational capabilities.

The chatbot leverages advanced NLP techniques to retrieve relevant information from the uploaded documents and generate accurate responses based on the context.

---

🚀 Features

• Document Upload: Users can upload their PDF documents for the chatbot to process.
• FAISS RAG: Utilizes FAISS for efficient similarity search and retrieval of relevant document chunks.
• LangChain Integration: Enhances the chatbot's ability to handle complex queries and generate context-aware responses.
• Portuguese Language Support: The chatbot understands and responds to queries in Portuguese.
• Interactive Interface: A user-friendly interface for seamless interaction with the chatbot.

---

📂 How to Install

Prerequisites

• Python 3.8 or higher
• pip (Python package manager)

Installation Steps

1. Clone the repository:

   git clone https://github.com/Jpzinn654/chatbot-documents

2. Install the required dependencies:

   pip install -r requirements.txt
   Set up the environment variables:
   Create a .env file in the root directory.

3. Run the application:

   cd src
   streamlit run main.py

• Open your browser and navigate to http://localhost:8501 to access the chatbot interface.

🛠️ How to Use

Upload a PDF Document:

• Click on the "Carregue PDF" button and select your document, after this click on "Processar", and wait to process

• Type your question in Portuguese in the chat input box, The chatbot will retrieve relevant information from the document and provide a response.

• Continue asking questions or request summaries, explanations, or specific details from the document.

📄 Technologies Used

• FAISS: For efficient similarity search and document retrieval.

• LangChain: For building the conversational AI pipeline.

• Streamlit: For the user-friendly web interface.

• Google/Flan-T5-Large: For advanced natural language understanding and generation.

🤝 Contributing

• Contributions are welcome! If you'd like to contribute, please follow these steps:

• Fork the repository, Create a new branch for your feature or bug fix, Commit your changes. Submit a pull request.

📜 License

• This project is licensed under the MIT License. See the LICENSE file for details.