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

Performance Vectors (and Confusion Matrix)

KrafNKrafN Member Posts: 13 Contributor II
edited November 2018 in Help
Hello everyone,

I have a process in which I train a classifier and test its performance afterwards. My example set has a binominal label and when I output the data generated by the binominal performance evaluator I get a nice view with information about (in this case) accuracy, precision and recall.

My problem is that the wrong label value is treated as positive class so is there an easy way to change that? Or is there an easy way to access the elements of the confusion matrix?

Answers

  • landland RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 2,529   Unicorn
    Hi,
    just use the remap binominals operator to define which value should be positive and which should be treated as negative.

    Greetings,
    Β  Sebastian
  • KrafNKrafN Member Posts: 13 Contributor II
    First of all, thanks for your advice.

    I also noticed that I can read the desired sample counts from the Performance Vector if I set the right flags in the Performance Evaluator Operator beforehand.
Sign In or Register to comment.