Decision Tree

funnyhatfunnyhat Member Posts: 13 Contributor II
edited November 2018 in Help
Hi, there.

I am a beginner of RM 5, I met a very strange thing in decision tree. my data have 468 examples, 5 regular attributes (gender, weight , length etc),2 special (one for id , one for label (true, false)), around 70 of them are true, when i try to use decision tree ,  only a false label shows, nothing else,  but the log said successfully, however if i test the golf or iris data, the results are correct.  I have no missing value.  anyone have any idea about it? is it possible related to impossible to make a cutoff?



  • Options
    landland RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 2,531 Unicorn
    this might be true, yes. The RapidMiner Decision tree performs two ways of pruning the tree: Before and after making splits. Might be it stops to early or prunes the tree back to the root. You should deactivate the pruning to test if the tree grows :)
    If yes, you could play around with the pruning related parameters to prevent pruning from cutting the hole tree.

  • Options
    funnyhatfunnyhat Member Posts: 13 Contributor II
    Thanks, I think that's the point...
Sign In or Register to comment.