real-estate-price-predictor

A Machine Learning + Flask Web App to predict house prices in Bangalore based on location, square footage, number of bedrooms (BHK), and bathrooms.

---

Project Overview

• Dataset: Bengaluru House Price Dataset (bengaluru_house_prices.csv)
• Model: Linear Regression (trained with data preprocessing and feature engineering)
• Backend: Python Flask API (server.py, util.py)
• Frontend: HTML, CSS, JavaScript (app.html, app.css, app.js)
• Deployment Ready: Can be hosted on Heroku / Render / AWS

---

Project Structure

├── artifacts/
│   ├── banglore_home_prices_model.pickle   # Trained ML model
│   ├── columns.json                        # Model features
│
├── data/
│   ├── bengaluru_house_prices.csv          # Raw dataset
│   ├── bhp.csv                             # Cleaned dataset
│
├── notebooks/
│   └── banglore_home_prices_final.ipynb    # Jupyter notebook for EDA & training
│
├── app.html                                # Frontend UI
├── app.css                                 # Styling
├── app.js                                  # Client-side logic
├── server.py                               # Flask server
├── util.py                                 # Helper functions
├── requirements.txt                        # Dependencies
└── README.md                               # Project Documentation

---

How It Works

1. User enters:
   • Area (sqft)
   • BHK
   • Bathrooms
   • Location
2. The request is sent to Flask backend (/predict_home_price).
3. The trained ML model predicts the price.
4. Result is displayed on the webpage in Lakhs (₹).

---

Installation & Setup

1. Clone the repository:

   git clone https://github.com/YOUR_USERNAME/bangalore-house-price-prediction.git
   cd bangalore-house-price-prediction

2. Create a virtual environment & activate it:

   python -m venv venv
   source venv/bin/activate    # Linux/Mac
   venv\Scripts\activate       # Windows

3. Install dependencies:

   pip install -r requirements.txt

4. Run the Flask server:

   python server.py

5. Open app.html in your browser → enter details → get predictions

---

Model Training

• Preprocessing: Removed outliers, handled missing values, averaged total_sqft ranges
• Feature Engineering: One-hot encoding for 241 Bangalore locations
• Model: Linear Regression (or XGBoost with 87.41% accuracy in testing)
• Saved: As banglore_home_prices_model.pickle

---

Screenshots

---

Future Improvements

• Integrate advanced ML models (e.g., Random Forest, XGBoost, CatBoost)
• Deploy live on Heroku/Render
• Expand dataset to include other Indian cities
• Enhance UI/UX with visualizations

---

Author

Prins Ambaliya

GitHub: PrinsAmbaliya

LinkedIn: https://www.linkedin.com/in/prins-ambaliya-bb7546367