image

🎉 🎉 RAPIDMINER 9.10 IS OUT!!! 🎉🎉

Download the latest version helping analytics teams accelerate time-to-value for streaming and IIOT use cases.

CLICK HERE TO DOWNLOAD

Apply Threshold to Polynomial Confusion Matrix

michaelglovenmichaelgloven RapidMiner Certified Analyst, Member Posts: 46  Guru
edited December 2018 in Help

Hi,

 

I have a polynomial attribute as a label and would like to improve my class recall of each label value (there are 9) by changing the threshold value used against the confidence values to make the prediction. Is it possible to use thresholds in this case or are threshold just applicable to binomial lables?

 

Answers

  • mschmitzmschmitz Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,053  RM Data Scientist

    Hi!

     

    The current threshold system is only for binominal labels. You can get a similar solution with Generate Attributes though. The problem with a polynominal threshold is that they might not be explicit.

     

    Have you tried Drop Uncertain? It sets predictions below a threshold to missing.

     

    Best,

    Martin

    - Head of Data Science Services at RapidMiner -
    Dortmund, Germany
Sign In or Register to comment.