draw.dendrogram3d: Draws a dendrogram in 3d

Description

Draw a dendrogram in 3d given clustering results and leaf positions. Labels at leaf tips may be specified

Usage

draw.dendrogram3d(cluster, positions, direction = c(0, 0, -1), scale = NULL,
heights = NULL, labels = NULL, label.colors = NULL, label.size = 3)

Arguments

cluster

Hierarchical clustering result of type hclust to be plotted.

positions

Positions (in 3d) of leaf tips. If 2d positions are given, these are placed in the z=0 plane.

direction

A vector specifying the direction in which the leaves point

scale

Maximum hieght of dendrogram

heights

A vector of heights, which may be used to over-ride the height information included in cluster (which are used by default if this is NULL)

labels

Vector of text labels to be placed at leaf tips.

label.colors

Vector of colors used used for labels.

label.size

Text size for labels.

Value

This function is called for the side effect it produces. It returns the id number of the last object drawn.

Details

May be used to validate 2D embedding results with those of cluster analysis. When cluster results and the corresponding 2D embedding are specified, this function places the positions in a rotatable environment with the cluster analysis result superposed on it. This allows the user to understand the relationship between the clustering and embedded results. Labels (or desired colors) may be placed at the leaf tips.

Examples

Run this code

mtcars.nMDS<-nMDS(mtcars,metric="euclidean")
mtcars.cluster<-hclust(dist(mtcars),method="complete")
draw.dendrogram3d(mtcars.cluster,mtcars.nMDS$x,labels=rownames(mtcars),label.size=0.5)

Run the code above in your browser using DataLab

Data Engineering and BI courses are free this week!