kmeans.getk {fdm2id}

Estimation of the number of clusters for K-means

Description

Estimate the optimal number of cluster of the K-means clustering method.

Usage

kmeans.getk(
  d,
  max = 9,
  criterion = "pseudo-F",
  graph = TRUE,
  nstart = 10,
  seed = NULL
)

Arguments

d

The dataset (matrix or data.frame).

max

The maximum number of clusters. Values from 2 to max are evaluated.

criterion

The criterion to be optimized. "pseudo-F" is the only criterion implemented in the current version.

graph

A logical indicating whether or not a graphic should be plotted.

nstart

The number of random sets chosen for kmeans initialization.

seed

A specified seed for random number generation.

Value

The optimal number of cluster of the K-means clustering method according to the chosen criterion.

See Also

pseudoF, kmeans

Examples

require (datasets)
data (iris)
kmeans.getk (iris [, -5])