Title: Cascade generalisation for ordinal problems

Authors: Sotiris Kotsiantis, Dimitris Kanellopoulos

Addresses: Department of Mathematics, University of Patras, GR 265 00, Patras, Greece. ' Educational Software Development Laboratory (ESDLab), Department of Mathematics, University of Patras, GR 265 00, Patras, Greece

Abstract: Given an ordered class, a researcher is not only interested in minimising the classification error, but also in minimising the distances between the actual and the predicted class. This paper offers an organised study of the various methodologies that have tried to handle this problem and presents an experimental study of these methodologies with a proposed cascade generalisation technique, which combines the predictions of a classification tree and a model tree algorithm. The paper concludes that the proposed technique can be a more robust solution to the problem since it minimises the distance between the actual and the predicted class and improves the classification accuracy.

Keywords: data mining; supervised machine learning; ranking learning; cascade generalisation; ordinal problems; classification tree; model tree; classification accuracy; classification errors.

DOI: 10.1504/IJAISC.2010.032512

International Journal of Artificial Intelligence and Soft Computing, 2010 Vol.2 No.1/2, pp.46 - 57

Published online: 04 Apr 2010 *

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