Soft margin trees

Abstract

From a multi-class learning task, in addition to a classi er, it is possible to infer some useful knowledge about the relationship between the classes involved. In this paper we propose a method to learn a hierarchical clustering of the set of classes. The usefulness of such clusterings has been exploited in bio-medical applications to nd out relations between diseases or populations of animals. The method proposed here de nes a distance between classes based on the margin maximization principle, and then builds the hierarchy using a linkage procedure. Moreover, to quantify the goodness of the hierarchies we de ne a measure. Finally, we present a set of experiments comparing the scores achieved by our approach with other methods