제어로봇시스템학회:학술대회논문집

Institute of Control, Robotics and Systems (제어로봇시스템학회)
권호 표지

Application of genetic algorithms to cluster analysis

  • Tagami, Takanori (Department of Information Science and Intelligent Systems, Faculty of Engineering, University of Tokushima) ;
  • Miyamoto, Sadaaki (Department of Information Science and Intelligent Systems, Faculty of Engineering, University of Tokushima) ;
  • Mogami, Yoshio (Department of Information Science and Intelligent Systems, Faculty of Engineering, University of Tokushima)
  • Published : 1993.10.01
Download PDF
⟨ Previous Next ⟩

Abstract

The aim of the present paper is to show the effectiveness of Genetic Algorithm for data classification problems in which the classification criteria are not the Euclidean distance. In particular, in order to improve a search performance of Genetic Algorithm, we introduce a concept of the degree of population diversity, and propose construction of genetic operators and the method of calculation for the fitness of an individual using the degree of population diversity. Then, we investigate their performances through numerical simulations.

Keywords