fast_dca: Discriminant Correspondence Analysis

In svs: Tools for Semantic Vector Spaces

Description

A fast procedure for computing discriminant correspondence analysis.

Usage

fast_dca(dat, clusters1 = NULL, clusters2 = NULL, members = FALSE)

Arguments

dat	Input data: can be a table or a data frame (but the data frame must have only two columns).
clusters1	A clustering of the first set of levels: either a list or the output of kmeans.
clusters2	A clustering of the second set of levels: either a list or the output of kmeans.
members	Logical indicating whether the (supplementary) coordinates for the individual levels should also be computed.

Value

A list with components:

val	The eigenvalues or principal inertias, indicating how much each latent axis explains.
cen1	The coordinates of the cluster centers for the first set of levels.
cen2	The coordinates of the cluster centers for the second set of levels.
mem1	If members = TRUE: The coordinates of the first set of individual levels.
mem2	If members = TRUE: The coordinates of the second set of individual levels.

References

Abdi, H. (2007) Discriminant correspondence analysis. In: N. Salkind (ed.) Encyclopedia of measurement and statistics. Thousand Oaks: SAGE.

Examples

SndT_Fra <- read.table(system.file("extdata", "SndT_Fra.txt", package = "svs"),
   header = TRUE, sep = "\t", quote = "\"", encoding = "UTF-8",
   stringsAsFactors = FALSE)
sca.SndT_Fra <- fast_sca(SndT_Fra)
kcl.SndT_Fra <- kmeans(sca.SndT_Fra$pos1, centers = 7)
dca.SndT_Fra <- fast_dca(SndT_Fra, clusters1 = kcl.SndT_Fra)
dca.SndT_Fra

svs documentation built on Nov. 10, 2020, 1:09 a.m.