An AI-powered web application that predicts Bengaluru house prices using Machine Learning. This project leverages Linear Regression, One-Hot Encoding, and an interactive Streamlit dashboard to estimate property prices based on user inputs such as location, total area, bedrooms, and bathrooms.
Real estate pricing is influenced by several factors including location, property size, and amenities. Estimating the correct market value manually can be difficult.
This project builds a Machine Learning model trained on the Bengaluru House Price Dataset to predict house prices instantly through a modern web interface.
The application follows a complete Data Science workflow:
- Data Collection
- Data Cleaning
- Exploratory Data Analysis
- Feature Engineering
- Machine Learning Model Training
- Interactive Visualization
- Web Deployment using Streamlit
✅ Predict Bengaluru House Prices
✅ AI-powered Property Valuation
✅ Interactive Streamlit Dashboard
✅ Beautiful Modern UI
✅ Real-time Price Prediction
✅ Property Summary
✅ Market Insights
✅ Price Distribution Analysis
✅ Area vs Price Visualization
✅ BHK Analysis
✅ Top Locations Analysis
✅ Download Prediction Report
✅ Prediction History
Dataset Name
Bengaluru House Price Dataset
The dataset contains thousands of Bengaluru property listings with information such as:
- Property Location
- Total Square Feet
- Number of Bedrooms
- Number of Bathrooms
- Property Price
- Area Information
The dataset undergoes several preprocessing steps before training.
- Removed unnecessary columns
- Removed missing values
- Converted area ranges into numeric values
- Extracted Bedrooms from Size column
- Removed invalid records
- Created Bedrooms feature
- Converted total_sqft into numerical values
- Grouped rare locations into "Other"
- Removed outliers using percentile method
- Location
- Total Square Feet
- Bathrooms
- Bedrooms
Target Variable
- Price (Lakhs ₹)
The project uses
combined with
using a Scikit-Learn Pipeline.
Pipeline
OneHotEncoder
↓
Linear Regression
↓
Price Prediction
Dataset
│
▼
Data Cleaning
│
▼
Feature Engineering
│
▼
Outlier Removal
│
▼
Train Test Split
│
▼
One Hot Encoding
│
▼
Linear Regression
│
▼
Model Evaluation
│
▼
Prediction
│
▼
Streamlit Deployment
- Python
- Pandas
- NumPy
- Scikit-Learn
- Category Encoders
- Plotly
- Streamlit
House-Price-Prediction/
│── app.py
│── Bengaluru_House_Data.csv
│── House Price Prediction Model.ipynb
│── requirements.txt
│── README.md
git clone https://github.com/shreya975/House-Price-Prediction.gitcd House-Price-Predictionpip install -r requirements.txtstreamlit run app.py(Add Screenshot Here)
(Add Screenshot Here)
(Add Screenshot Here)
The application includes
- Price Distribution
- Area vs Price Scatter Plot
- BHK Distribution
- Top Locations
- Feature Importance
- Property Score
- AI Confidence Meter
- Affordability Meter
Users provide
- 📍 Location
- 📐 Total Square Feet
- 🛏 Bedrooms
- 🛁 Bathrooms
The application predicts
🏡 Estimated House Price (₹ Lakhs)
- Random Forest Regression
- XGBoost Regression
- LightGBM
- Deep Learning Models
- Google Maps Integration
- Property Recommendation System
- User Login
- Database Support
- API Deployment
- Mobile Responsive Version
The application is deployed using Streamlit Cloud
👉 https://house-price-prediction-cbrveafrrwmzxsneuttr6u.streamlit.app/