TITLE:
Estimating the number of data clusters via the contrast statistic
AUTHORS:
Yuriy Lyakh, Vitaliy Gurianov, Oleg Gorshkov, Yuriy Vihovanets
KEYWORDS:
SOM Neural Network; Clustering; Gap Statistic; Silhouette Statistic
JOURNAL NAME:
Journal of Biomedical Science and Engineering,
Vol.5 No.2,
February
28,
2012
ABSTRACT: A new method (the Contrast statistic) for estimating the number of clusters in a set of data is proposed. The technique uses the output of self-organising map clustering algorithm, comparing the change in dependency of “Contrast” value upon clusters number to that expected under a uniform distribution. A simulation study shows that the Contrast statistic can be used successfully either, when variables describing the object in a multi-dimensional space are independent (ideal objects) or dependent (real biological objects).