Variables importance Simple Regression Tree Learner - Need help!

Hello all,

I use Knime for a school assignment.
I have made a process with a Simple Regression Tree Learner and I was currious if there is something that can show me the Variables importance of my data set.
someone who can help?

Hi,
I don’t know if it is a good idea to determine variable importance from a single tree. Generally, you can use the Regression Tree View of the Simple Regression Tree Learner to have a look at the tree. Variables that are used for a split near the root of the tree are more likely to be important, but it would be better to train a Random Forest and have a look at the attribute statistics this outputs. Here you can see which attributes were used how often for a 1st, 2nd, and 3rd level split and derive an importance measure from that.
Kind regards
Alexander
Screenshot 2020-03-27 at 13.19.06

3 Likes

This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.