RapidMiner 9.8 Beta is now available

Be one of the first to get your hands on the new features. More details and downloads here:

GET RAPIDMINER 9.8 BETA

Please help me about polynominal label problem.

sukanyadedkaew1sukanyadedkaew1 Member Posts: 17 Contributor II




Answers

  • BalazsBaranyBalazsBarany Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert Posts: 474   Unicorn
    Hi,

    learners have different capabilities. For example, Logistic Regression is only able to decide between two classes (true and false, 0 and 1, ...).

    However, there is a meta-learner that goes through all classes one by one and calculates confidences for every "class vs. all the others". Then it selects the class with the highest confidence. The operator is called "Polynominal by Binominal Classification" and it's a wrapper around the learner, in your case the Logistic Regression. Just put it into the process instead of the Regression and put Logistic Regression inside it.

    Regards,
    Balázs


    sukanyadedkaew1
  • sukanyadedkaew1sukanyadedkaew1 Member Posts: 17 Contributor II


    The result of logistic regression shows. Thanks a lot.

    Took
    mschmitz
Sign In or Register to comment.