clst-package: Classification by local similarity threshold

In clst: Classification by local similarity threshold

Description

Package for modified nearest-neighbor classification based on calculation of a similarity threshold distinguishing within-group from between-group comparisons.

Details

Package:	clst
Type:	Package
License:	GPL-3
Author:	Noah Hoffman <[e-mail]>

Index:

Further information is available in the following vignettes:

clstDemo	clst (source, pdf)

TODO: write package overview.

Author(s)

Noah Hoffman

Maintainer: <[e-mail]>

See Also

cmdscale

Examples

library(clst)
packageDescription("clst")
data(iris)
dmat <- as.matrix(dist(iris[,1:4], method="euclidean"))
groups <- iris$Species
i <- 1
cc <- classify(dmat, groups, dvect=dmat[i,])
cat('query at i =',i,'is species',paste('I.', groups[i]),'\n')
printClst(cc)
i <- 125
cc <- classify(dmat, groups, dvect=dmat[i,])
cat('query at i =',i,'is species',paste('I.', groups[i]),'\n')
printClst(cc)

clst documentation built on Nov. 8, 2020, 5:41 p.m.