testpart: Identify Misclassified Plots in a Partition

In optpart: Optimal Partitioning of Similarity Relations

Description

testopt analyzes the mean similarity of each sample to the cluster to which it is assigned to all other clusters, and lists those samples which have similarity higher to another cluster than to the one to which they are assigned.

Usage

testpart(part,ord=TRUE)

Arguments

part	a object of class ‘partana’ from partana or optpart
ord	a switch to control whether the output is ordered

Details

Simply examines each sample plot, comparing the mean similarity of that sample to all other samples in the cluster to which it is assigned as compared to its mean similarity to all other clusters. Samples which are more similar to other clusters than to the one to which they are assigned are listed in a table which gives their current cluster assignment, the cluster to which they are more similar, and the mean similarities of that sample to all clusters.

If ‘ord=TRUE’ then the output is ordered to reflect target clusters.

Value

a table of values

Author(s)

David W. Roberts [e-mail]

See Also

partana, partana, silhouette

Examples

data(shoshveg)
dis.bc <- dsvdis(shoshveg,'bray/curtis')
opt.5 <- optpart(5,dis.bc)
testpart(opt.5)

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