HUM calculator and HUM package for R: easy-to-use software tools for multicategory receiver operating characteristic analysis

Abstract

Receiver operating characteristic (ROC) analysis is usually applied in bioinformatics to evaluate the abilities of biological markers to differentiate between the presence or absence of a disease. It includes the derivation of the useful scalar performance measure area under the ROC curve for binary classification tasks. As real applications often deal with more than two classes, multicategory ROC analysis and the corresponding hypervolume under the manifold (HUM) measure have become a topic of growing interest. To support researchers in carrying out multicategory ROC analysis, we have developed two tools in different programming environments which feature user-friendly, object-oriented and flexible interfaces and enable the user to compute HUM values and plot 2D- and 3D-ROC curves. The software is freely available from our Web site http://public.ostfalia.de/∼klawonn/HUM.htm