KMEANS

k-means clustering

    [LABELS,A] = KMEANS(A,K,MAXIT,INIT,FID)

Input
A	Matrix or dataset
K	Number of clusters to be found (optional; default: 2)
MAXIT	maximum number of iterations (optional; default: 50)
INIT	Labels for initialisation, or  'rand' : take at random K objects as initial means, or  'kcentres' : use KCENTRES for initialisation (default)
FID	File ID to write progress to (default [], see PRPROGRESS)

Output
LABELS	Cluster assignments, 1..K
A	Dataset with original data and labels LABELS

Description

K-means clustering of data vectors in A.

See also

datasets, hclust, kcentres, modeseek, emclust, prprogress,