plot.pkbc {QuadratiK}R Documentation

Plotting method for Poisson kernel-based clustering

Description

Plots for a pkbc object.

Usage

## S4 method for signature 'pkbc,ANY'
plot(x, k = NULL, true_label = NULL, pca_res = FALSE, ...)

Arguments

x

Object of class pkbc

k

number of considered clusters. If it is not provided the scatter plot is displayed for each value of number of clusters present in the x object

true_label

factor or vector of true membership to clusters (if available). It must have the same length of final memberships.

pca_res

Logical. If TRUE the results from PCALocantore are also reported (when dimension is greater than 3).

...

Additional arguments that can be passed to the plot function

Details

Value

The scatter-plot(s) and the elbow plot.

Note

The elbow plot is commonly used as a graphical method for choosing the appropriate number of clusters. Specifically, plotting the wcss versus the number of clusters, the suggested number of clusters correspond to the point in which the plotted line has the greatest change in slope, showing an elbow.

References

Locantore, N., Marron, J.S., Simpson, D.G. et al. (1999) "Robust principal component analysis for functional data." Test 8, 1–73. https://doi.org/10.1007/BF02595862

See Also

pkbc() for the clustering algorithm
pkbc for the class object definition.

Examples

dat<-matrix(rnorm(300),ncol=3)
pkbc_res<- pkbc(dat, 3)
plot(pkbc_res, 3)


[Package QuadratiK version 1.1.2 Index]