Interpreting RapidMiner regression accuracy results

Casper72

Hi, forgive me for asking such a simple question, but is "class recall" actually how well a model performs in predicting classes? At first I thought "class precision" would provide this information, but when testing the models my results consistently add up to "class recall".

sgenzer

Hi @Casper72 there's a pretty good video on confusion matrices on the RapidMiner Academy:

https://academy.rapidminer.com/learn/video/introduction-to-performance-measurement

Scott

Casper72

Thank you so much!

Casper72

Update: the ordering of numbers I get in the confusion matrix does not correspond to what I see in the "Predictions" tab. When I manually calculate  class precision and class recall from predictions vs actual values, it looks like the confusion matrix has switched the values around(?) from:

TruePositive       FalsePositive
FalseNegative    TrueNegative

to:

TruePositive     FalseNegative    
FalsePositive   TrueNegative