calinhara {fpc} | R Documentation |
Calinski-Harabasz index for estimating the number of clusters,
based on an observations/variables-matrix here. A distance based
version is available through cluster.stats
.
calinhara(x,clustering,cn=max(clustering))
x |
data matrix or data frame. |
clustering |
vector of integers. Clustering. |
cn |
integer. Number of clusters. |
Calinski-Harabasz statistic, which is
(n-cn)*sum(diag(B))/((cn-1)*sum(diag(W)))
. B being the
between-cluster means,
and W being the within-clusters covariance matrix.
Christian Hennig chrish@stats.ucl.ac.uk http://www.homepages.ucl.ac.uk/~ucakche
Calinski, R. B., and Harabasz, J. (1974) A Dendrite Method for Cluster Analysis, Communications in Statistics, 3, 1-27.
set.seed(98765) iriss <- iris[sample(150,20),-5] km <- kmeans(iriss,3) calinhara(iriss,km$cluster)