I have a dataset containing non-normally distributed variables that I want to feed into a distance-based classifier (e.g. K-means). Is it ok to just subtract the mean and divide by the standard deviation to "standardize" it? Or should some other transformation be performed? Is it even necessary to standardize the data before feeding it into such a classifier?

This is similar to this question, but the other question doesn't cover the k-means angle.

Edit: I would like to standardize the data to prevent one of my covariates from dominating the distance calculation. Some of my covariates are bounded between 0..200 and others between 0..2.

  • $\begingroup$ What does the standardization need to achieve? $\endgroup$ – Glen_b -Reinstate Monica Sep 27 '13 at 0:05
  • $\begingroup$ I think you answered your own question. You should standardize the data if your inputs have different scales. $\endgroup$ – Akavall Sep 30 '13 at 2:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.