This is a complete Data Science + NLP-powered project focused on Mumbai’s residential property market. It consists of three powerful modules: price prediction, advanced analytics, and a smart chatbot — all tied together with a modern Streamlit frontend.

The Mumbai real estate market is vast and highly dynamic. Buyers often struggle to:

• Get accurate price estimates
• Explore meaningful trends (like price by locality, age, BHK)
• Find answers to their questions quickly

We propose an intelligent real estate system that:

• Predicts property prices using machine learning
• Visualizes important property patterns and trends
• Answers user queries using an AI chatbot trained on real data

All these components are unified in a clean, interactive Streamlit dashboard.

We used mumbaipropdataset.csv, a structured dataset containing:

• Property details like AREA, BHK, FACING, AGE, PRICE, DESCRIPTION, LAT, LONG, etc.
• Listings from across Mumbai residential sectors

• Built using Random Forest Regressor (n=500)
• Achieved 91% accuracy
• Input: Property features like area, locality, BHK, age, etc.
• Output: Predicted property price

• Built with Plotly inside Streamlit
• Shows:
  • Scatter plots: Area vs Price
  • Sunburst: City → Locality → BHK
  • Radial charts: Price by Facing
  • Line charts: Price vs Age
  • Word cloud from descriptions
  • Pie charts: BHK filtered by locality

• Powered by a lightweight LLM/chatbot interface
• Trained over mumbaipropdataset.csv
• Handles both Q&A and visualization queries
• Hosted on Colab with GPU, integrated with Streamlit via an API

git clone https://github.com/your-username/your-repo.git
cd your-repo

• Open RealEstateChatbot.ipynb on Colab (T4 GPU)
• Run all cells — an API link will be generated
• Copy this API URL

• Go to:

  pages/RealtEase.py

• Paste the API URL at line 12

  url = "https://your-colab-api.ngrok-free.app"

️⚠️ Note: The chatbot will keep working until your Colab session is live.

1. Create your environment (e.g., dslab) and install dependencies:

pip install virtualenv

virtualenv dslab

cd dslab/Scripts && activate

2. Install the required packages:

pip install -r requirements.txt

3. Launch the Streamlit app:

streamlit run Real_Estate_Project.py

🎉 That's it! Your Mumbai Real Estate Intelligence App is now live.

• Replace RF with XGBoost or CatBoost
• Improve chatbot with LLM fine-tuning or LangChain
• Deploy chatbot via WhatsApp/Telegram

• Guide: Ms. Sonal Balpande
• Institute: A.P. Shah Institute of Technology, Mumbai
• Built as part of the academic mini project (TE IT - 2024-25)

This project is for academic and research purposes only.