Options

Sample (Stratified)

tobybtobyb Member Posts: 11 Contributor II
edited June 2019 in Help
It would be nice if the current operator was modified so that the proportion of classes can be specified.  I have worked with several datasets where one class is underrepresented.  As a result the model classifies everything as the class that is overrepresented.  The modified operator I am proposing would be able to capture a number of cases by class.  This way you could control how many cases were sampled from each class.  This type of preprocesing in my experience leads to meaningful models.

Answers

  • Options
    tobybtobyb Member Posts: 11 Contributor II
    I am replying to my own post.  I am new to this forum and I posted this request before I looked thru the other posts.  I realized after the fact that Axel has proposed something similar in the Balancing Operator post.  I apologize for duplicating this request.
  • Options
    landland RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 2,531 Unicorn
    Hi,
    at least you showed the urgency of this proposal :)

    Greetings,
      Sebastian
Sign In or Register to comment.