Regression project project as a part of Data&AI training at BeCode
To create a Machine Learning Regression project based on real data for Belgian housing market.
Our application will build various types of Regression models and predict approximate prices for properties in Belgium. This Machine Learning application is based on Streamlit library of Python. It gives the user the possibility to choose between Brussels, Flanders and Wallonia regions and houses, apartments, offices and industry property types. Outliers removal option and hyperparameter tuning for regression models are key feature of our project application. For performance analysys, a comparative study between Linear Regression, K-Nearest Neighbor Regression and Random Forest Regression models is presented on basis of error between predictions and real data.
Click on Belgian Housing Market Prediction App
- Clone the repo
- Install the required libraries
We used the data scraped from the real estate website (Immoweb) for more than 12,000 properties. The data was then preprocessed to remove duplicates and none values. Exploratory Data Analysis was done and on basis of correlation we selected 'Area' as the feature to determine our target the 'Price'.
Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.
- Fork the Project
- Create your Feature Branch (
git checkout -b feature/AmazingFeature) - Commit your Changes (
git commit -m 'Add some AmazingFeature') - Push to the Branch (
git push origin feature/AmazingFeature) - Open a Pull Request
Distributed under the MIT License.
Please, contact any of the authors via GitHub.