JoaoAlmeida-dev / MachineLearning-Project

1 stars 0 forks source link

Merging/Dropping features #11

Open JoaoAlmeida-dev opened 2 years ago

JoaoAlmeida-dev commented 2 years ago

First of all, I am dropping a variable if two variables are highly correlated. To find these correlated pairs, I have used the correlations matrix again (see section 6.1). For instance: GarageCars and GarageArea have a correlation of 0.89. Of those two, I am dropping the variable with the lowest correlation with SalePrice (which is GarageArea with a SalePrice correlation of 0.62. GarageCars has a SalePrice correlation of 0.64).