Inductive Genetic Programming with Decision Trees

Article type: Research Article

Authors: Nikolaev, Nikolay I.^{a; *} | Slavov, Vanio^{b; 1}

Affiliations: [a] Department of Computer Science, American University in Bulgaria, Blagoevgrad 2700, Bulgaria | [b] Information Technologies Lab, New Bulgarian University, Sofia 1113, Bulgaria

Correspondence: [*] Corresponding author. E-mail: nikolaev@nws.aubg.bg.

Note: [1] E-mail: vslavov@inf.nbu.acad.bg.

Abstract: This article proposes a study of inductive Genetic Programming with Decision Trees (GPDT). The theoretical underpinning is an approach to the development of fitness functions for improving the search guidance. The approach relies on analysis of the global fitness landscape structure with a statistical correlation measure. The basic idea is that the fitness landscape could be made informative enough to enable efficient search navigation. We demonstrate that by a careful design of the fitness function the global landscape becomes smoother, its correlation increases, and facilitates the search. Another claim is that the fitness function has not only to mitigate navigation difficulties, but also to guarantee maintenance of decision trees with low syntactic complexity and high predictive accuracy.

Keywords: Inductive concept learning, Decision trees, Genetic programming, Fitness landscapes

DOI: 10.3233/IDA-1998-2104

Journal: Intelligent Data Analysis, vol. 2, no. 1, pp. 31-44, 1998