Fig. 3
From: Efficiency of different measures for defining the applicability domain of classification models
Ames data set employing classification random forests. Receiver operating characteristic (ROC) curves for the classification technique random forest in combination with the Ames data set. The results are shown for the confidence measures and the two novelty measures \(\cos_{\alpha }\) and \(\gamma_{Euc}\). The inferior performance of the novelty measures can easily be seen. For intermediately difficult problems the impact of a well performing AD measure is largest (for details see text)