sg-house is a Linear Regression project created by NUS CEG students Marcus Ong and Ng Andre, to predict housing prices in Singapore, based on past datasets.
Using the HDB resale price dataset obtained in 2019, sg-house applies linear regression, a form of supervised machine learning, in order to investigate the relationship between HDB prices against the given criteria, and provide a reliable prediction of the housing prices.
Current criteria used:
- Floor Area in square metres
- Age of house
- Years of lease remaining
Ensure that the dataset hdb_resale_price_Aug2019.csv is in the same directory as sg-house.py.
As the program employs certain python3 libraries, check that the following packages are installed on your machine before running:
- numpy
- pandas
- matplotlib
- seaborn
- scikit-learn
> pip3 list
> pip3 install numpy
> pip3 install pandas
> pip3 install matplotlib
> pip3 install seaborn
> pip3 install scikit-learn
usage: python3 sg-house.py
Note: The matplotlib plot may not appear if using WSL terminal for execution. Alternatively, run the code via GUI, such as Visual Studio Code or by double clicking the app.
input:
Choose your town:
What is the floor_area_sqm
What is the age
What is the remaining_lease_years
output:
Your estimated house price is
As the app is still in its development stage, we will be improving on it. Current plans include:
- Distance to primary schools in vicinity
- Distance to nearest MRT
- Improvement on data visualisation
- Scrape data directly from data.gov.sg for updated housing prices
- Distance to shopping centres and other amenities