xx_recolor: Recolor Classification IDs

In EMCluster: EM Algorithm for Model-Based Clustering of Finite Mixture Gaussian Distribution

Recolor Classification IDs

Description

These functions return new classification IDs.

Usage

recolor(id.target, id.class, scatter.class = NULL, scatter.target = NULL)
rematch(tg.id, cl.id)
recode(id)

Arguments

id.target	target class ids.
id.class	original class ids.
scatter.class	scatter class ids.
scatter.target	scatter target class ids.
id	class ids.
tg.id	target class ids.
cl.id	class ids.

Details

The function recolor colors id.target in accordance with the most likely candidate in id.class. Note that if scatter is present, then the class given by 0 is represented as scatter and it is assumed to be the same for both classifications.

The function rematch returns a list as id.trcl and id.prcl. It is the heart of the recolor function and is usually called from recolor.

The function recode reoders classes to eliminate group ids without any members. It is assumed that the group ids are integers.

Value

See Details.

Author(s)

Ranjan Maitra.

References

https://www.stat.iastate.edu/people/ranjan-maitra

Examples

## Not run: 
library(EMCluster, quietly = TRUE)

true.id <- c(1, 1, 1, 2, 2, 2, 3, 3, 3)
pred.id <- c(2, 1, 2, 1, 1, 1, 2, 1, 1)

recolor(pred.id, true.id)

## End(Not run)

EMCluster documentation built on Sept. 11, 2024, 5:27 p.m.