PamParam-class: Partitioning around medoids

In LTLA/bluster: Clustering Algorithms for Bioconductor

Partitioning around medoids

Description

Partition observations into k-medoids as a more robust version of k-means.

Usage

PamParam(
  centers,
  metric = NULL,
  medoids = NULL,
  nstart = NULL,
  stand = NULL,
  do.swap = NULL,
  variant = NULL
)

## S4 method for signature 'ANY,PamParam'
clusterRows(x, BLUSPARAM, full = FALSE)

Arguments

centers	An integer scalar specifying the number of centers. Alternatively, a function that takes the number of observations and returns the number of centers.
metric, medoids, nstart, stand, do.swap, variant	Further arguments to pass to pam. Set to the function defaults if not supplied.
x	A numeric matrix-like object where rows represent observations and columns represent variables.
BLUSPARAM	A PamParam object.
full	Logical scalar indicating whether the full PAM statistics should be returned.

Details

This class usually requires the user to specify the number of clusters beforehand. However, we can also allow the number of clusters to vary as a function of the number of observations. The latter is occasionally useful, e.g., to allow the clustering to automatically become more granular for large datasets.

To modify an existing PamParam object x, users can simply call x[[i]] or x[[i]] <- value where i is any argument used in the constructor.

Value

The PamParam constructor will return a PamParam object with the specified parameters.

The clusterRows method will return a factor of length equal to nrow(x) containing the cluster assignments. If full=TRUE, a list is returned with clusters (the factor, as above) and objects (a list containing pam, the direct output of pam).

Author(s)

Aaron Lun

See Also

pam, which actually does all the heavy lifting.

KmeansParam, for the more commonly used k-means algorithm.

ClaraParam, for a scalable extension to the PAM approach.

Examples

clusterRows(iris[,1:4], PamParam(centers=4))
clusterRows(iris[,1:4], PamParam(centers=4, variant="faster", do.swap=FALSE))
clusterRows(iris[,1:4], PamParam(centers=sqrt))

LTLA/bluster documentation built on Aug. 20, 2023, 5:39 a.m.