What is the best way to encode features when clustering data?

Question

I have a dataset with numerical and categorical features. I am trying to run a k-means algorithm to find clusters of data.

What is the best way to encode categorical features?

I have been doing one hot encoding but I guess that it can be easily improved.

score 0 · Answer 1 · edited Dec 19 '19 at 22:06

0

Try Hierarchical Clustering, Kmeans will not serve the purpose if you have categorical data.

edited Dec 19 '19 at 22:06

Stephen Rauch

answered Dec 19 '19 at 21:06

Syenix

For what I have seen in the other post one of the proposed ideas is to use K-Modes algorithm. – Carlos Mougan Dec 20 '19 at 11:30

1 Answers1