While determining the house price I don't want to throw away geo location as feature. At least I want to see if there is correlation. Geo-location is more like categorical variable. Is there a way to make it useful feature
-
$\begingroup$ head over to kaggle's NYC Taxi Comp $\endgroup$ – Aditya May 12 '18 at 18:07
-
$\begingroup$ Which type of model(s) are you using in your analysis? Linear regression and decision tree models can work with categorical variables as well. $\endgroup$ – raghu May 13 '18 at 4:42
$\begingroup$
$\endgroup$
Why is it categorical? You can use the longitude / latitude numeric coordinates as two features.