umatrix: U-Matrix

In yasomi: Yet Another Self Organising Map Implementation

Description

Plot a U-matrix for a Self-Organising Map.

Usage

umatrix(som, ...)

Arguments

som	an object of class "som"
...	additional parameters transmitted to the low level plot function plot.sompdist

Details

Many variations of Ultsch and Siemon's original U-matrix have been used in scientific publications. The umatrix function provides one of the simplest solution, based on a coloring of the glyphs used to represent the prior structure of the fitted SOM. With the default parameters, the colour of each cell encodes the mean distance between the corresponding prototype and its direct neighbours (using heat.colors). Details can be found in plot.sompdist.

Value

umatrix silently returns the object of class "sompdist" build from the som parameters and used to construct the graphical representation.

Author(s)

Fabrice Rossi

References

Ultsch, A. and Siemon, H. P. (1990) Kohonen's self organizing feature maps for exploratory data analysis, in: Proceedings of International Neural Network Conference (INNC'90).

See Also

See plot.sompdist for details, distance.grid for possibly smoother plots.

Examples

data(iris)
# scaling
data <- scale(iris[1:4])

# a medium hexagonal grid
sg <- somgrid(xdim=15,ydim=15,topo="hex")

# choose a good SOM via Kaski and Lagus' error measure
st <- som.tune(data,sg,som.tunecontrol(sg,criterion=error.kaskilagus))
som <- st$best.som

# display the umatrix
umatrix(som)

yasomi documentation built on May 2, 2019, 5:59 p.m.