Title: Bootstrap a Clustering Solution to Establish the Stability of the Clusters
Version: 1.2.2
Description: Providing a cluster allocation for n samples, either with an $n \times p$ data matrix or an $n \times n$ distance matrix, a bootstrap procedure is performed. The proportion of bootstrap replicates where a pair of samples cluster in the same cluster indicates who tightly the samples in a particular cluster clusters together.
License: MIT + file LICENSE
Encoding: UTF-8
RoxygenNote: 7.2.3
Depends: R (≥ 2.10)
LazyData: true
Imports: graphics, grDevices, stats
Suggests: fpc, testthat (≥ 3.0.0)
Config/testthat/edition: 3
NeedsCompilation: no
Packaged: 2024-07-08 11:19:46 UTC; Sugnet
Author: Sugnet Lubbe ORCID iD [aut, cre, cph]
Maintainer: Sugnet Lubbe <slubbe@sun.ac.za>
Repository: CRAN
Date/Publication: 2024-07-08 11:30:02 UTC

Heatmap of the proportion of bootstrap replicates where objects cluster together

Description

Heatmap of the proportion of bootstrap replicates where objects cluster together

Usage

boot.proportions(
  x,
  col = grDevices::heat.colors(101, rev = TRUE),
  show.vals = F,
  text.col = "black",
  cluster.col = "firebrick",
  ...
)

Arguments

x

an object of class clusboot

col

vector of colours for shading to indicate proportion values

show.vals

logical value indicating whether proportion values should be added to individual cells

text.col

colour of text for show.vals if TRUE

cluster.col

colour of lines demarcating cluster membership

...

more arguments to be passed to plot()

Examples

out <- clusboot (scale(case.study.psychiatrist), B=100, k=6, clustering.func=complete.linkage)
boot.proportions(out)

Produces silhouette plots

Description

Produces silhouette plots

Usage

boot.silhouette(clusboot.out, ...)

Arguments

clusboot.out

an object of class clusboot

...

more arguments to be passed to barplot()

Value

list of silhouette widths

Examples

out <- clusboot (scale(case.study.psychiatrist), B=100, k=6, clustering.func=complete.linkage)
boot.silhouette(out)

Computes the silhouette vales based on proportion of times items cluster together

Description

Computes the silhouette vales based on proportion of times items cluster together

Usage

calc.silhouette(clusboot.out)

Arguments

clusboot.out

an object of class clusboot

Value

an object of class clusboot

Patient by psychiatric symptom data

Description

Presence/absence ratings of 24 psychiatric symptoms in 30 psychiatric inpatients made by an individual psychiatrist. The data have been collected in a case study of an individual psychiatrist to identify his implicit taxonomy.

Usage

case.study.psychiatrist

Format

case.study.psychiatrist

A data frame with 30 observations on the following 28 variables:

V1

inappropriate affect, appearance or behavior; binary vector

V2

interview belligerence - negativism; binary vector

V3

agitation - excitement; binary vector

V4

retardation; binary vector

V5

lack of emotions; binary vector

V6

speech disorganization; binary vector

V7

grandiosity; binary vector

V8

suspicion - ideas of persecution; binary vector

V9

hallucinations - delusions; binary vector

V10

overt anger; binary vector

V11

depression; binary vector

V12

anxiety; binary vector

V13

obsession - compulsion; binary vector

V14

suicide; binary vector

V15

self injury; binary vector

V16

somatic concerns; binary vector

V17

social isolation; binary vector

V18

daily routine impairment; binary vector

V19

leisure time impairment; binary vector

V20

antisocial impulses or acts; binary vector

V21

alcohol abuse; binary vector

V22

drug abuse; binary vector

V23

disorientation; binary vector

V24

memory impairment; binary vector

V25

rating on Global Assessment Scale, a 101-point scale for overall severity of psychiatric disturbance; a numeric vector

V26

Affective (Affective Disorder or Anxiety Disorder); binary vector

V27

Psychotic (Schizophrenic Disorder or Paranoid Disorder); binary vector

V28

Substance abuse (Substance Use Disorder or Substance-Induced Disorder); binary vector

Details

The data set forms part of the International Federation of Classification Societies Cluster Benchmark Data Repository

Source

Van Mechelen, I., & De Boeck, P. (1989). Implicit taxonomy in psychiatric diagnosis: A case study. Journal of Social and Clinical Psychology, 8, 276-287. https://ifcs.boku.ac.at/repository/data/case_study_psychiatrist/index.html

Performs bootstrap on a cluster analysis output

Description

Performs bootstrap on a cluster analysis output

Usage

clusboot(datmat, B = 1000, clustering.func = complete.linkage, ...)

Arguments

datmat

a data matrix or distance object which will be the input to the clustering function

B

number of bootstrap replicates

clustering.func

the function which will perform the clustering and output a vector of cluster memberships

...

more arguments to be passed to the clustering function

Details

Any R function performing cluster analysis can be specified in clustering.func although a wrapper function is typically needed to isolate only the vector output of cluster memberships. See ?complete.linkage as an example. Should users perfer to use alternative resamling schemes, other than the bootstrap, Hennig (2007) discuss a variety of options which could be accessed by specifying clustering.func = fpc.clusterboot. In addition, the sampling method is specified in the argument bootmethod and additional arguments for the function clusterboot in the package fpc must be given. Note that only the resampling facilities of clusterboot is utilised while the computation of proportions and silhouette widths remain unchanged. The output object of class clusboot will remain unchanged as only the resampling section of clusterboot is used.

Value

an object of class clusboot which is a list with the following components:

proportions

matrix of size nxn with cell ij containing the proportion of bootstrap replicates in which object i and object j clustered together.

clustering

a vector of length n containing the cluster membership of the n input objects.

sil

a vector of length the number of clusters containing the bootstrap-silhouette values for the clusters.

indv.sil

a vector of length n containing the bootstrap-silhouette values for the individual objects.

sil.order

a vector of length n containing the ordering of the n objects used by the functions boot.silhouette and boot.proportions to order objects in the same cluster adjacent and clusters in decreasing order of cluster tightness.

ave.sil.width

the overall stability of the clustering solution, obtained by averaging over the individual object bootstrap-silhouette values.

References

Hennig, C., 2007. Cluster-wise assessment of cluster stability. Computational Statistics & Data Analysis, 52(1), pp.258-271. Lubbe, S., 2024. Bootstrapping Cluster Analysis Solutions with the R Package ClusBoot. Austrian Journal of Statistics, 53(3), pp.1-19.

Examples

clusboot (scale(case.study.psychiatrist), B=100, k=6, clustering.func=complete.linkage)
library(fpc)
clusboot (scale(case.study.psychiatrist), B=100, k=6, clustering.func=fpc.clusterboot,
          clustermethod=hclustCBI, method="complete", bootmethod="subset", subtuning=10)

Wrapper function for performing complete linkage clustering

Description

Wrapper function for performing complete linkage clustering

Usage

complete.linkage(X, k)

Arguments

X

samples x variables data matrix

k

number of clusters

Value

vector of cluster memberships

Examples

complete.linkage(scale(case.study.psychiatrist), k=6)

Resampling according to the methods discussed in Hennig (2007)

Description

Resampling according to the methods discussed in Hennig (2007)

Usage

fpc.clusterboot(
  data,
  B,
  distances = (inherits(data, "dist")),
  bootmethod = "boot",
  bscompare = TRUE,
  multipleboot = FALSE,
  jittertuning = 0.05,
  noisetuning = c(0.05, 4),
  subtuning = floor(nrow(data)/2),
  clustermethod,
  noisemethod = FALSE,
  count = TRUE,
  seed = NULL,
  datatomatrix = TRUE,
  ...
)

Arguments

data

a data matrix or distance object which will be the input to the clustering function

B

number of bootstrap replicates

distances

see ?fpc::clusterboot

bootmethod

see ?fpc::clusterboot

bscompare

see ?fpc::clusterboot

multipleboot

see ?fpc::clusterboot

jittertuning

see ?fpc::clusterboot

noisetuning

see ?fpc::clusterboot

subtuning

see ?fpc::clusterboot

clustermethod

see ?fpc::clusterboot

noisemethod

see ?fpc::clusterboot

count

see ?fpc::clusterboot

seed

see ?fpc::clusterboot

datatomatrix

see ?fpc::clusterboot

...

additional arguments to be sent to the function specified in clustermethod

Value

a list with two components; boot.out contains the computations for clusboot and out contains the clustering solution of the original data set

MDS plot of similarities given by the proportion of bootstrap replicates where objects cluster together

Description

MDS plot of similarities given by the proportion of bootstrap replicates where objects cluster together

Usage

## S3 method for class 'clusboot'
plot(x, col, show.silhouette = TRUE, ...)

Arguments

x

an object of class clusboot

col

single colour or a vector specifying a colour for each object

show.silhouette

logical indicating whether plotting character size should represent the inidivual silhouette values

...

more arguments to be passed to plot()

Value

matrix of similarities (proportions)

Examples

out <- clusboot (scale(case.study.psychiatrist), B=100, k=6, clustering.func=complete.linkage)
plot(out)