plot.divclust: Dendrogram with binary questions

In chavent/divclust: Divisive hierarchical clustering

Description

Plot the dendrogram produced by divclust algorithm with the binary questions associated with each split. The number of binary questions drawn on the dendrogram can be changed and if the text of the question is to long, is is printed on the console.

Usage

## S3 method for class 'divclust'
plot(x, nqbin = 3, label = TRUE, ...)

Arguments

x	object of class divclust
nqbin	an integer between 0 and K-1 (K is the number of leaves in the tree). Indicates the number of binary questions drawn on plot. This parameter is used to plot the binary questions only of the top levels of the dendrogram. The default value is 3.
label	If TRUE, the labels of the observations are drawn.
...	further arguments passed from other methods

See Also

divclust

Examples

data(protein) # pure quantitatives data
c_tot <- divclust(protein) # full clustering
plot(c_tot)
plot(c_tot,nqbin=4) # the text of the 4th question is printed in the console
c_4 <- divclust(protein, K=4) # stops the clustering to 4 clusters
plot(c_4)

data(dogs) # pure qualitative data
c_tot <- divclust(dogs) # full clustering
plot(c_tot,nqbin=4) # the text of the 4th question is printed in the console

data(wine) # mixed data
data <- wine[,1:29]
c_tot <- divclust(data) # full clustering
plot(c_tot)

chavent/divclust documentation built on May 13, 2019, 3:38 p.m.