2
$\begingroup$

What is the good way working with 'Age' attribute? Don't touch it or should it be scaled? Below photo shows my results 'Before' and 'After' standardization.

enter image description here

$\endgroup$
4
  • 2
    $\begingroup$ The answer depends on what you are using the attribute for. Input to neural network? Input to linear model? If linear, do you expect the model to be interpretable? $\endgroup$
    – noe
    Commented Jan 22, 2023 at 17:27
  • $\begingroup$ My attribute is from dataset, this dataset is used for learning a machine learning model (s), like simple ones. And I'm not sure if the model is going to be interpretable. $\endgroup$ Commented Jan 22, 2023 at 17:31
  • 2
    $\begingroup$ Different scales of the dependent variables affect the interpretability of feature importance in linear models, that's why you normally transform them into the same value scale. Therefore, if you are going to use it in a logistic regression where you need interpretability. $\endgroup$
    – noe
    Commented Jan 22, 2023 at 17:47
  • $\begingroup$ does standardization in your case (graphs) change the meaning of data ? what is the formula you apply for it? $\endgroup$ Commented Feb 19, 2023 at 3:25

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.