numero.evaluate: Self-organizing map statistics

In Numero: Statistical Framework to Define Subgroups in Complex Datasets

Self-organizing map statistics

Description

Evaluate regional variation of data values on a self-organizing map

Usage

numero.evaluate(model, data, ranked = TRUE, n = 1000)

Arguments

model	A list object that contains a self-organizing map and a data layout.
data	A matrix or a data frame.
ranked	If true, a rank transform is applied to avoid problems from skewed distributions or outliers.
n	Maximum number of permutations per data column.

Details

The input argument model can be the output from numero.create() or from numero.quality().

Value

A list with named elements: som contains the self-organizing map, layout contains the district assignments for data points, planes contains smoothed district averages from nroAggregate(), the element ranges contains the reference ranges to be used in nroColorize(), the element statistics contains the output from nroPermute(), the element palette is the name of the colormap and the element data contains the data points that were used for calculating the statistics.

Examples

# Import data.
fname <- system.file("extdata", "finndiane.txt", package = "Numero")
dataset <- read.delim(file = fname)

# Set identities and manage missing data.
dataset <- numero.clean(dataset, identity = "INDEX")

# Prepare training variables.
trvars <- c("CHOL", "HDL2C", "TG", "CREAT", "uALB")
trdata <- numero.prepare(data = dataset, variables = trvars)

# Create a self-organizing map.
sm <- numero.create(data = trdata)
qc <- numero.quality(model = sm)

# Evaluate map statistics.
results <- numero.evaluate(model = qc, data = dataset)
print(results$statistics[,c("TRAINING", "Z", "P.z", "P.freq")])

Numero documentation built on Jan. 9, 2023, 9:08 a.m.