The hyperellipsoidal clustering using genetic algorithm

Author

Song, Wang ; Feng, Ma ; Wei, Shi ; Shaowei, Xia

Author_Institution

Dept. of Autom., Tsinghua Univ., Beijing, China

Volume

1

fYear

1997

fDate

28-31 Oct 1997

Firstpage

592

Abstract

Hyperellipsoidal clustering can characterize the distribution of the clusters better than common hyperspherical clustering. In this paper, it is proved that the direct application of Mahalanobis distance instead of Euclidean distance as the similarity measure cannot acquire the hyperellipsoidal clustering. Based on the analysis a new similarity measure suitable to hyperellipsoidal clustering is presented and a genetic algorithm is applied to optimize the modified clustering cost function. The simulation experiments show the efficiency of the new algorithm

Keywords

genetic algorithms; pattern recognition; search problems; Euclidean distance; Mahalanobis distance; cluster distribution; clustering cost function; genetic algorithm; hyperellipsoidal clustering; hyperspherical clustering; pattern recognition; similarity measure; simulation experiments; Algorithm design and analysis; Automation; Clustering algorithms; Computational modeling; Cost function; Covariance matrix; Euclidean distance; Genetic algorithms; Machine learning algorithms; Shape;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Processing Systems, 1997. ICIPS '97. 1997 IEEE International Conference on

Conference_Location

Beijing

Print_ISBN

0-7803-4253-4

Type

conf

DOI

10.1109/ICIPS.1997.672853

Filename

672853