A modified version of the K-means algorithm with a distance based on cluster symmetry

Authors
Su, MS Chou, CH
Citation
Ms. Su et Ch. Chou, A modified version of the K-means algorithm with a distance based on cluster symmetry, IEEE PATT A, 23(6), 2001, pp. 674-680
Citations number
24
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
ISSN journal
01628828 → ACNP
Volume
23
Issue
6
Year of publication
2001
Pages
674 - 680
Database
ISI
SICI code
0162-8828(200106)23:6<674:AMVOTK>2.0.ZU;2-0
Abstract
In this paper, we propose a modified version of the K-means algorithm to cl uster data. The proposed algorithm adopts a novel nonmetric distance measur e based on the idea of "point symmetry." This kind of "point symmetry dista nce" can be applied in data clustering and human face detection. Several da ta sets are used to illustrate its effectiveness.