RapidMiner 9.7 is Now Available

Lots of amazing new improvements including true version control! Learn more about what's new here.

CLICK HERE TO DOWNLOAD

"Using create threshold with logistic regression and forward selection"

hmhsinghmhsing Member Posts: 21  Maven
edited June 2019 in Help

I am using forward selection and logistic regression with create threshold, when I put 0.8 in threshold, true in first class and false in second class, the prediction result, however shows true when confidence(false) is greater than 0.8. Is this somthing wrong? Thanks!

 

Tagged:

Best Answer

  • earmijoearmijo Posts: 263   Unicorn
    Solution Accepted

    RM is behaving exactly the way it is supposed to behave. 

     

    The rule is, given two classes A (let's call it "first class") and B ("second class"):

     

    Classify the observation as class B when confidence(B)>threshold,

    classify it as class A otherwise. 

     

    Short answer: Flip the roles. Put "true" in the "second class" spot. 

Sign In or Register to comment.