• Introduction
• Installation
• Usage
• Results
• Conclusion

This project is an end-to-end conversational real estate database retrieval system that leverages Streamlit for the user interface, Google Palm LLM for natural language understanding, and the Langchain framework for managing conversational flow. The system interacts with a MySQL database to provide dynamic Q&A, using few-shot learning and Hugging Face embeddings stored in ChromaDB to enhance query accuracy and relevance.

To set up this project, follow these steps:

1. Clone the repository:

   git clone https://github.com/nanmaharaj/EstateMate-QueryBot.git
   cd conversational-real-estate-retrieval

2. Install the required dependencies:

   pip install -r requirements.txt

3. Set up the database:

   • Ensure you have a MySQL database set up with your real estate data.
   • You can import data using the provided SQL file.

1. Start the Streamlit UI:

   streamlit run main.py

2. Enter your query in English: Use the Streamlit interface to enter your real estate-related queries.

The conversational system successfully retrieves and responds to queries about real estate data with high accuracy. Key features include:

• Natural language understanding powered by Google Palm LLM.
• Conversational management using Langchain.
• Enhanced query accuracy with few-shot learning and Hugging Face embeddings.

This project demonstrates the integration of advanced AI technologies to create a dynamic and interactive real estate query system. The combination of Google Palm LLM, Langchain, and Hugging Face embeddings offers a robust solution for conversational data retrieval.