Housing Price Prediction

A Flask-based web application that predicts house prices based on user input. The application uses a trained machine learning model to make predictions.

Features

• Predicts house prices based on various input features such as area, bedrooms, bathrooms, and more.
• Machine learning pipeline with preprocessing, feature engineering, and model selection.
• User-friendly interface built with Flask and HTML/CSS.

Dataset

Source: Kaggle Housing Prices Dataset
Description: This dataset contains features such as area, bedrooms, bathrooms, and other attributes of houses, along with their respective prices.

Machine Learning Model

The application uses a LightGBM Regressor, selected after training and evaluating the following models:

• Linear Regression
• Ridge Regression
• Lasso Regression
• Random Forest Regressor
• LightGBM Regressor

Preprocessing Pipeline

• Numerical Features: Scaled using MinMaxScaler.
• Categorical Features:
  • Ordinal features encoded using OrdinalEncoder.
  • Nominal features encoded using OneHotEncoder.

Model Selection Process

• A Pipeline was used to combine preprocessing and model training steps.
• Hyperparameter tuning was performed using GridSearchCV with 5-fold cross-validation.
• Metrics used for evaluation included:
  • Mean Absolute Error (MAE)
  • Mean Squared Error (MSE)
  • Root Mean Squared Error (RMSE)
  • Mean Squared Logarithmic Error (MSLE)

Results

• The LightGBM Regressor was the best-performing model with optimized hyperparameters.
• The model achieved an R² score of 0.6 on the test dataset.

Installation

1. Clone the repository:

   git clone https://github.com/spicyneutrino/Housing-Prediction.git
   cd Housing-Prediction

2. Create a virtual environment:

   python3 -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install the required dependencies:

   pip install -r requirements.txt

4. Run the application:

   flask run

   Open your browser and go to http://127.0.0.1:5000/.

Demo

Below is a demo of the Flask application in action:

Deployment

This project is suitable for deployment on Heroku.

Future Improvements

• Enhanced User Interface: Improve the web design for a more modern and intuitive experience.
• Additional Features: Incorporate more house attributes such as proximity to schools or public transport.
• Model Expansion: Experiment with other advanced models or ensemble techniques for improved accuracy.
• API Integration: Develop APIs to allow external applications to interact with the prediction service.
• Cloud Deployment: Deploy the application on other platforms like AWS or Azure for scalability.

Acknowledgments

• Dataset: Kaggle Housing Prices Dataset
• Technologies: Flask, Python, and Scikit-Learn.
• Machine learning model tuned with LightGBM and GridSearchCV.