Vector decision trees

Article type: Research Article

Authors: Šprogar, Matej | Kokol, Peter | Babič, Špela Hleb | Podgorelec, Vili | Zorman, Milan

Affiliations: University of Maribor, Laboratory for system design, Smetanova 17, 2000 Maribor, Slovenia. E-mail: matej.sprogar@uni-mb.si, kokol@uni-mb.si

Abstract: The article presents the extension of a common decision tree concept to a multidimensional -- vector -- decision tree constructed with the help of evolutionary techniques. In contrary to the common decision tree the vector decision tree can make more than just one suggestion per input sample. It has the functionality of many separate decision trees acting on a same set of training data and answering different questions. Vector decision tree is therefore simple in its form, is easy to use and analyze and can express some relationships between decisions not visible before. To explore and test the possibilities of this concept we developed a software tool -- DecRain -- for building vector decision trees using the ideas of evolutionary computing. Generated vector decision trees showed good results in comparison to classical decision trees. The concept of vector decision trees can be safely and effectively used in any decision making process.

Keywords: decision trees, decision-making, data mining, genetic algorithms, machine learning

DOI: 10.3233/IDA-2000-43-410

Journal: Intelligent Data Analysis, vol. 4, no. 3-4, pp. 305-321, 2000