clusterCompactness: cluster compactness measures

In nguforche/UnsupRF: Unsupervised Random Forest Clustering

Description

Calculate 5 within cluster compactness or homogenerity measures

Usage

clusterCompactness(distance = NULL, clusters, Data = NULL,
  method = "euclidean", ensemble = TRUE, compact.measure = "SW", ...)

Arguments

distance	dissimilarity matrix
clusters	existing cluster partitions to determine compactness
ensemble	whether to find the optimal compactness measure by majority rule
compact.measure	if ensemble=FALSE, provide the compactness measure to return. Options are DM cluster diabmeter SW Cluster Silhouette width WD Within cluster distance gap Cluster gap
...	further arguments passed to or from other methods.
data	data matrix from which cluster partitions were derived. Note that one of distance or data must be provided

Value

the numeric compactness measure

nguforche/UnsupRF documentation built on May 5, 2019, 4:51 p.m.