Learning to Classify Ordinal Data: The Data Replication Method

摘要

Classification of ordinal data is one of the most important tasks ofrelation learning. This paper introduces a new machine learningparadigm specifically intended for classification problems where theclasses have a natural order. The technique reduces the problem ofclassifying ordered classes to the standard two-class problem. Theintroduced method is then mapped into support vector machines andneural networks. Generalization bounds of the proposed ordinalclassifier are also provided. An experimental study with artificial andreal data sets, including an application to gene expression analysis,verifies the usefulness of the proposed approach. color="gray">