How to calculate inter-model correlation for an ensemble
Dear experts,
I have read that empirically, stacking ensemble tend to yield better results when there is a significant diversity among the models, one way to examine that is looking for a low correlation between the predictions of the classifier.
In R there is function modelCor (caret) which can be used to check it, Is there any same function/method/operator in Rapidminer ?
Thank you very much for your help and response.