An improvement of the NEC criterion for assessing the number of clusters in a mixture model

Authors
Biernacki, C Celeux, G Govaert, G
Citation
C. Biernacki et al., An improvement of the NEC criterion for assessing the number of clusters in a mixture model, PATT REC L, 20(3), 1999, pp. 267-272
Citations number
5
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
PATTERN RECOGNITION LETTERS
ISSN journal
01678655 → ACNP
Volume
20
Issue
3
Year of publication
1999
Pages
267 - 272
Database
ISI
SICI code
0167-8655(199903)20:3<267:AIOTNC>2.0.ZU;2-C
Abstract
The entropy criterion NEC showed good performances for choosing the number of clusters arising from a mixture model. But it was not valid to decide be tween one and more than one clusters. This note presents a natural extensio n of this criterion to deal with this situation. Illustrative experiments e xhibit good behavior of this modified entropy criterion. (C) 1999 Elsevier Science B.V. All rights reserved.