Continuing my journey, I arrived at the next test of my burgeoning data science capabilities, a dataset containing close to 22,000 homes in Seattle. The data contained the price of each home, along with other descriptive features, such as the square footage and number of bedrooms or bathrooms. My objective was to find the best predictors for the price of these homes in the Greater Seattle area.
Does the location of a home affect it’s price?
I began by looking at the geographical data associated with the homes, to see if there was a correlation between location and price.
KeplerGL, a mapping tool used by Uber, provided a great platform for this geospatial analysis. With just a few lines of code in Jupyter, I was able to work with a HTML map within the notebook.
Breaking down house prices by quantile allowed me to see that the more affluent homes were primarily concentrated around the Bellevue area.
