AUC with cost-sensitive classifier, why does it change?
it is unclear to me why in a dataset like breast-cancer.arff, using
RandomForest in a cost-sensitive way (cost of Minimize expected cost = True,
provides me an AUC of *0.498* whereas a standard random forest provides me
an AUC of *0.634*. Since the cost-sensitive classifier should just apply
different thresholds on results, and since AUC is threshold-independent, I
do not understand why the two AUC are different.