The characterization of classification problems by classifier disagreements

In this paper we try to characterize a set of classification problems. For this, we use the disagreement between a set of standard classifiers. The disagreement patterns do not only point towards different types of classification problems, but also indicate the novelty and the usefulness of a classifier with respect to a set of classification problems and classifiers. Some experiments show when known classification problems become unknown after changing their feature size or their training set size.