Decision tree - One Branch to rule them all

Ben_SuenBen_Suen Member Posts: 3 Learner I
edited June 2020 in Help
Hey guys, 

I've been working on a decision tree for a few days now, while we are still figuring out how to best generate it one pattern we are seeing is that almost every time one 'branch' leading to a particular leaf ends up completely dominating the tree resulting in almost 99% of the values being able to be predicted by that branch. 

Is this normal? Should I be correcting/changing the data in some way to counter act this?

We have tried the Auto Model function as well and are running into the same issue. 

Answers

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,453 RM Data Scientist
    Hi,
    well.. this sometimes happens. I would'nt worry to much if the prediction accuracy is good.

    What you can do is change the criterion. I know that these kind of balanced trees happen if you use accuracy as a criterion.

    Best,
    Martin
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
    sgenzer
Sign In or Register to comment.