Next: Introduction

The Case for Genetic Algorithms in Fuzzy Clustering

L.O. Hall¹, B. Ozyurt¹, J.C. Bezdek²
¹Department of Computer Science and Engineering
University of South Florida
Tampa, Fl. 33620
²Department of Computer Science
University of West Florida
Pensacola, Fl. 32514

Abstract:

Genetic algorithms provide a method of doing optimization. This paper examines how genetic algorithms can be used in performing fuzzy clustering of data. In this paper the effectiveness of the genetic algorithm in optimizing the fuzzy and hard c-means objective function is illustrated empirically. Factoring in the time cost of a genetic algorithm limits the case for genetic algorithms to domains with many extrema or experiments with new, untried objective functions.

Introduction
Clustering with HCM and FCM
Genetically guided clustering
Data set descriptions
Experiments and results
- HCM experiments
Time considerations
Summary
References
About this document ...

Larry Hall
5/26/1998