daltoolbox (version 1.3.767)

cluster_kmeans: k-means

Description

k-means clustering using stats::kmeans.

Usage

cluster_kmeans(k = 1)

Value

returns a k-means object.

Arguments

k

the number of clusters to form.

Details

Partitions data into k clusters minimizing within‑cluster sum of squares. The intrinsic quality metric returned is the total within‑cluster SSE (lower is better).

The base clusterer() uses wcss as a generic default, but cluster_kmeans() specializes that choice because k-means is usually interpreted through compactness and separation between centroid-based groups.

Default evaluation in cluster_kmeans() is:

• main metric: metric_silhouette()

• internal evaluation: silhouette, davies_bouldin, calinski_harabasz

• external evaluation: entropy, purity, adjusted_rand_index

The general wcss fallback is not kept as the main metric here because it decreases mechanically as k grows and is therefore weaker as a standalone quality criterion for comparing partitions.

References

MacQueen, J. (1967). Some Methods for classification and Analysis of Multivariate Observations. Lloyd, S. (1982). Least squares quantization in PCM. Rousseeuw, P. J. (1987). Silhouettes: A graphical aid to the interpretation and validation of cluster analysis. Davies, D. L., & Bouldin, D. W. (1979). A cluster separation measure. Calinski, T., & Harabasz, J. (1974). A dendrite method for cluster analysis.

Examples

# setup clustering
model <- cluster_kmeans(k=3)

#load dataset
data(iris)

# build model
model <- fit(model, iris[,1:4])
clu <- cluster(model, iris[,1:4])
table(clu)

# evaluate model using external metric
eval <- evaluate(model, clu, iris$Species)
eval