A Machine Learning web application that predicts the estimated price of a house based on property details such as area, bedrooms, city, locality, amenities, and nearby facilities.
The project is built using Python, Scikit-learn, and Gradio, and is deployed as an interactive web application.
📌 Features
Predict house prices using Machine Learning
User-friendly Gradio web interface
Supports multiple cities and localities
Takes property details and amenities as input
Uses a trained Random Forest Regression model
Ready for deployment on Hugging Face Spaces
📊 Dataset
The dataset contains 10,000 house records with features including:
City
Locality
Property Type
Area (sq.ft.)
Bedrooms
Bathrooms
Floors
Age of House
Parking
Furnishing
Facing
Balcony
Garden
Swimming Pool
Gym
Lift
Security
Distance to City Center
School Distance
Hospital Distance
Metro Distance
Crime Index
Pollution Index
Target Variable
House Price
🤖 Machine Learning Models Compared
The following regression algorithms were trained and evaluated:
Model
R² Score
Random Forest Regressor
0.8193 ✅
Gradient Boosting Regressor
0.8056
Linear Regression
0.7190
Decision Tree Regressor
0.5487
The Random Forest Regressor achieved the best performance and was selected as the final model.
