If a dataset has a variable with categorical values (A, B, C, D, E, F), would converting them to numeric and breaking them down into separate columns (for example, column A that will have 1s for all occurrences of A and 0s for non-occurrences), would that affect variability?

If yes, how would a problem like that can be handled? (assuming conversion to numeric values and standardization has been done)

  • $\begingroup$ It seems you're asking about dimensionality reduction with categorical variables after one-hot encoding, but I'm not sure I understand specifically what you're asking (e.g. variability of what? and handle what problem?). Could you elaborate? $\endgroup$ – user20160 Oct 11 '18 at 5:15
  • $\begingroup$ @user20160 in other words, the variability of the categorical column - will it still remain after breaking up categories into different columns? $\endgroup$ – Feyzi Bagirov Oct 12 '18 at 3:00

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.