California Housing Prices Prediction

A machine learning project to predict median house values in California using the classic California Housing dataset. Achieved 82.38% accuracy with Random Forest Regressor.

📊 Project Overview

This project uses the California Housing Prices dataset to build and evaluate a regression model for predicting median house prices based on features like location, rooms, population, and income.

Current Model Performance:
R² Score: 0.8238380642630709 (82.38% variance explained)

🛠 Tech Stack

• Language: Python
• Notebook: Jupyter Notebook (house.ipynb)
• Libraries: pandas, numpy, scikit-learn, matplotlib/seaborn
• IDE: Visual Studio Code
• Dataset: Kaggle California Housing Prices

📁 Project Structure

california-housing-prediction/
│
├── housing.csv              # Dataset (20,640 samples, 10 features)
├── house.ipynb             # Main analysis & ML pipeline
└── README.md              # Project documentation

🏠 Dataset

Source: California Housing Prices (Kaggle)

Features (10 total):

• MedInc: Median income in block group
• HouseAge: Median house age in block
• AveRooms: Average number of rooms per household
• AveBedrms: Average number of bedrooms per household
• Population: Block population
• AveOccup: Average number of household members
• Latitude, Longitude: Block group location
• ocean_proximity: Proximity to ocean (categorical)
• median_house_value: Target (median house value in $100,000s)

Dataset Size: 20,640 samples × 10 features

🚀 Quick Start

1. Clone/Download the repository
2. Install dependencies:

   pip install pandas numpy scikit-learn matplotlib seaborn jupyter

3. Open house.ipynb in VS Code or Jupyter
4. Run all cells to see the complete pipeline
5. View results - R² score of 0.8238 on test set!

📈 Model Performance

Metric	Score
R² Score (Test Set)	0.8238
Status	✅ Implemented up to forest.score(x_test_s, y_test)

Model: Random Forest Regressor

🔄 Workflow Implemented

1. ✅ Data Loading - housing.csv
2. ✅ EDA - Statistical analysis & visualizations
3. ✅ Preprocessing - Handling categorical variables, scaling
4. ✅ Train-Test Split
5. ✅ Random Forest Model Training
6. ✅ Model Evaluation - R² score: 0.8238

📊 Key Results

Random Forest R² Score: 0.8238380642630709

The model explains 82.38% of the variance in house prices!

🎯 Next Steps (Planned Enhancements)

• Hyperparameter tuning (GridSearchCV/RandomizedSearchCV)
• Cross-validation
• Feature importance analysis & visualization
• Compare with other models (XGBoost, LightGBM, Neural Networks)
• Model deployment (Flask/FastAPI)
• Interactive dashboard (Streamlit/Plotly Dash)

📝 Usage

Open house.ipynb and run all cells sequentially. The notebook contains:

1. Complete data preprocessing pipeline
2. Exploratory data analysis with visualizations
3. Model training and evaluation
4. Performance metrics and interpretation

🤝 Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

📄 License

This project is open source and available under the MIT License.

🙏 Acknowledgments

• California Housing Prices Dataset by @camnugent
• Scikit-learn documentation & examples
• YouTube Tutorial for initial inspiration

---

Built with ❤️ by Avani Parab | First-year CSE(AI&ML) Student

---