plot.ewkm: Plot Entropy Weighted K-Means Weights

In wskm: Weighted k-Means Clustering

Description

Plot a heatmap showing the variable weights from the subspace clustering.

Usage

## S3 method for class 'ewkm'
plot(x, ...)
## S3 method for class 'ewkm'
levelplot(x, ...)

Arguments

x	an object of class ewkm.
...	arguments passed on through to heatmap.

Details

The entopy weighted k-means clustering algorithm is a subspace clusterer ideal for high dimensional data. Along with each cluster we also obtain variable weights that provide a relative measure of the importance of each variable to that cluster.

This plot visualises these relative measures of variable importance for each of the clusters using a heatmap. The top dendrogram highlights the relationship between the clusters and the right side dendrogram provides a visual clue to the correlation between the variables.

The plot.ewkm() function uses heatmap() to display the weights. The levelplot.ewkm() uses levelplot() with dendrogramGlobs from the lattice package. Note that plot() will immediately draw the plot while levelplot() does not draw immediately but returns a result object which must be plot()ed.

Author(s)

Graham J Williams

Examples

myewkm <- ewkm(iris[1:4], 3, lambda=0.5, maxiter=100)

plot(myewkm)

wskm documentation built on April 5, 2020, 1:07 a.m.