Skip to contents

Summarize cluster labels

summary.dpmixgpd_cluster_labels.Rd

Summarize a representative clustering for training data or new observations.

Usage

# S3 method for class 'dpmixgpd_cluster_labels'
summary(object, top_n = 5L, order_by = c("size", "label"), vars = NULL, ...)

Arguments

object

Cluster labels object.

top_n

Number of populated clusters to profile when attached data are available.

order_by

Ordering rule for descriptive cluster profiles:

  • "size": decreasing cluster size

  • "label": ascending cluster label

vars

Optional character vector of numeric columns to summarize within each cluster.

...

Unused.

Value

Summary list containing cluster sizes, optional cluster-level descriptive summaries, and, when available, assignment-certainty summaries.

Details

If score or probability matrices are attached, certainty is summarized by the rowwise maxima \(\max_k p_{ik}\), which quantify how strongly each observation is assigned to its selected cluster. When the labels object also carries attached training or prediction data, the summary includes descriptive mean/sd profiles for the first populated clusters.

See also

predict.dpmixgpd_cluster_fit(), plot.dpmixgpd_cluster_labels(), summary.dpmixgpd_cluster_fit().

Other cluster workflow: dpmgpd.cluster(), dpmix.cluster(), plot.dpmixgpd_cluster_bundle(), plot.dpmixgpd_cluster_fit(), plot.dpmixgpd_cluster_labels(), plot.dpmixgpd_cluster_psm(), predict.dpmixgpd_cluster_fit(), print.dpmixgpd_cluster_bundle(), print.dpmixgpd_cluster_fit(), print.dpmixgpd_cluster_labels(), print.dpmixgpd_cluster_psm(), summary.dpmixgpd_cluster_bundle(), summary.dpmixgpd_cluster_fit(), summary.dpmixgpd_cluster_psm()