Finding the most relevant variables for classification models

[carlocelis]

How do we know which variables are the best predictors for classification models? Are there coefficients similar to linear regression?

[GuillaumeBot]

Hello @carlocelis,

there is two different things: the choice of number of components (or factor) and keeping only one question per component for the sake of clarity.

Of which variable are you talking? (can you give examples from the html presented in class?)

Cheers,

Guillaume

[carlocelis]

Hi Guillaume, I think you're talking about factoring and segmentation. My question is more on classification models such as ctree and rpart. Thanks!