If I have a feature with replicated values but each of these values indicates a different piece of information.

example: feature 'street name' with value 'A' which some of these 'A's are for Boston city and the others for New York City.

In this case, should I change the street name values which indicate NYC to something like 'A_NYC'? I assume that if I kept the 'A' values with no change then there will be a relation between Boston's 'A' and NYC and built a correlation between unrelated features.

I hope I made my point clear and sorry if this question was answered previously and I didn't notice.

  • $\begingroup$ Your intuition is correct. If you want streets to not relate when they comes from different towns a good solution is to apply that concatenation. $\endgroup$ – rapaio Oct 10 '19 at 5:19
  • $\begingroup$ @rapaio thank you for the help :) $\endgroup$ – Dia Abujaber Oct 10 '19 at 17:45

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.