sp.hclust: Hierarchical Clustering for Data on Hypersphere

In kisungyou/RiemSphere: What the package doesas

Description

Hierarchical Clustering for Data on Hypersphere

Usage

sp.hclust(
  x,
  type = c("intrinsic", "extrinsic"),
  method = c("single", "complete", "average", "mcquitty", "ward.D", "ward.D2",
    "centroid", "median"),
  members = NULL
)

Arguments

x	an (n\times p) row-stacked matrix for \mathbb{S}^{p-1}.
type	type of distance, either "intrinsic" or "extrinsic".
method	the agglomeration method to be used. This must be (an unambiguous abbreviation of) one of "single", "complete", "average", "mcquitty", "ward.D", "ward.D2", "centroid" or "median".
members	NULL or a vector whose length equals the number of observations. See hclust for details.

Value

an object of class hclust. See hclust for details.

Examples

## generate two-cluster data
mymu1 = c(0,0,0,1)  # center of class 1
mymu2 = c(-1,0,0,0) # center of class 2

x1 = rvmf(20, mymu1, kappa=5)
x2 = rvmf(20, mymu2, kappa=5)
xx = rbind(x1,x2)

## apply hierarchical clustering with different methods
hc1 <- sp.hclust(xx, method="single")
hc2 <- sp.hclust(xx, method="complete")
hc3 <- sp.hclust(xx, method="average")

## visualize
## Not run: 
opar <- par(mfrow=c(1,3), pty="s")
plot(hc1, main="'single'")
plot(hc2, main="'complete'")
plot(hc3, main="'average'")
par(opar)

## End(Not run)

kisungyou/RiemSphere documentation built on April 11, 2020, 4:04 a.m.