HCLUSTwrapper: Wrapper to do hierarchical clustering

In btaschler/mcap: Model-Based Clustering in High Dimensions via Adaptive Projections

Description

Hierarchical clustering using Euclidean distances and (by default) the Ward criterion. The wrapper can perform an optimisation over clustering methods in case true labels are provided (i.e. optimising cluster assignments w.r.t. the Rand index).

Usage

HCLUSTwrapper(xx, k, method = "ward.D", true_labels = NULL,
  verbose = FALSE)

Arguments

xx	The data matrix (n x p).
k	The number of clusters.
method	Clustering method (see hclust). If method = "all" and true_labels is provided: optimise clustering w.r.t. Rand index and return method with highest adjusted RI. Default: method = "Ward.D".
true_labels	Vector of true cluster assignments (when provided, it is used to compute the Rand index).
verbose	Logical, when true: print progress information.

Value

model_fit	Model fit (output of hclust).
aRI	Adjusted Rand index (when true_labels is provided).

Author(s)

Bernd Taschler: [e-mail]

See Also

hclust, adjustedRandIndex

Examples

  HCLUSTwrapper(xx = matrix(rnorm(500),50,10), k = 2)

btaschler/mcap documentation built on May 26, 2019, 1:31 a.m.