How did rapidminer Decision tree AUC being calculated

johnny5550822

Hi all,

How does rapdiminer decision tree (the basic one) AUC being calculated? For example, I have a dataset with only 1 parameter, and then I use 10 fold cross validation. The "performance" operator gives me AUCs (Optimistic, pessimistic, general), how did it generate? (I am guessing something like this: rapidminer try different cut-off for the single variable and determine the AUC?) ??? ???

Also, what's the difference among optimistic, pessimistic, and general ?

Thanks a lot!
Johnny

MariusHelf

Hi Johnny,

does this post help? http://rapid-i.com/rapidforum/index.php/topic,5698.msg20162.html#msg20162

Best regards,
Marius