Title: Is an ordinal class structure useful in classifier learning?

Authors: Jens C. Huhn, Eyke Hullermeier

Addresses: Department of Mathematics and Computer Science, Marburg University, Hans-Meerwein-Str., 35032 Marburg, Germany. ' Department of Mathematics and Computer Science, Marburg University, Hans-Meerwein-Str., 35032 Marburg, Germany

Abstract: In recent years, a number of machine learning algorithms have been developed for the problem of ordinal classification. These algorithms try to exploit, in one way or the other, the order information of the problem, essentially relying on the assumption that the ordinal structure of the set of class labels is also reflected in the topology of the instance space. The purpose of this paper is to investigate, on an experimental basis, the validity of this assumption. Moreover, we seek to answer the question to what extent existing techniques and learning algorithms for ordinal classification are able to exploit order information and which properties of these techniques are important in this regard.

Keywords: ordinal classification; binary decomposition; nested dichotomies; pairwise classification; class structure; classifier learning; machine learning.

DOI: 10.1504/IJDMMM.2008.022537

International Journal of Data Mining, Modelling and Management, 2008 Vol.1 No.1, pp.45 - 67

Published online: 14 Jan 2009 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article