reordclust: Re-order Clusters in a Classification

In optpart: Optimal Partitioning of Similarity Relations

Description

In it's simplest form simply reassigns cluster numbers in an existing classification to re-order tables and graphs. Can also be used to combine clusters into a fewer number of clusters.

Usage

reordclust(clustering,from,to)

Arguments

clustering	a vector of (interger) cluster mmeberships, or an object of class ‘clustering’, ‘partana’, or ‘partition’
from	an integer vector equal in length to the number of clusters that specifies the current clusters
to	an integer vector equal in length to the number of clusters that specifies the clusters the current clusters map to

Details

The function simply maps cluster numbers in the ‘from’ vector to the respective cluster number in the ‘to’ vector.

Value

an object of class ‘clustering’

Note

As demonstrated in the examples below, reordclass can also combine existing clusters into fewer clusters while reordering if more than one cluster in the ‘from’ vector maps to the same cluster in the ‘to’ cluster.

Author(s)

David W. Roberts [e-mail]

References

http://ecology.msu.montana.edu/labdsv/R

See Also

mergeclust

Examples

data(shoshveg)
dis.bc <- dsvdis(shoshveg,'bray')
opt.10 <- optpart(10,dis.bc)
new <- reordclust(opt.10,1:10,c(1,3,5,7,9,2,4,6,8,10))
           # simply re-ordered
new2 <- reordclust(opt.10,1:10,c(1,1,2,2,3,3,4,4,5,5))
           # merge 1 and 2 into 1, 3 and 4 into 2, etc.

optpart documentation built on March 26, 2020, 6:18 p.m.