Let’s say we have a data frame that looks like this: Df:

 A    B
0  45  78
1  5   34
2  3   0
3  56  0
4  34  0 
5  23  0

Does the fact that we have more values different than zero in column A would affect feature importance result, which means would this mean that A is more important, or it doesn’t matter ?

  • 1
    0 might have special meaning depending on the feature, as well as your task. – Ankit Seth Aug 10 at 11:41
  • so having so many 0 does not necessarily mean that the feature is less important, it's an information that the algorithm takes in consideration like any other number ? – Oumab10 Aug 10 at 11:47
  • Yes, that is correct. – Ankit Seth Aug 10 at 11:49

Your Answer

 
discard

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.