Numero (version 1.9.6)

nroPlot: Plot a self-organizing map

Description

Create a graphical interface for selecting subgroups from multiple map colorings simultaneously.

Usage

nroPlot(topology, colors, labels = NULL, subplot = NULL,
        interactive = FALSE, clear = NULL)
nroPlot.save(file, topology, colors, labels = NULL,
             subplot = NULL, font = 1.0)

Value

The main function returns a data frame with K rows that contains the topology and subgrouping information. The .save subroutine returns the number of bytes written in the output file.

Arguments

topology

A data frame with K rows and six or more columns that contain the district positions of a self-organizing map and optional region assignments.

colors

A character vector with K elements or a K x N matrix of hexadecimal color codes as strings.

labels

A character vector with K elements or a K x N matrix of district labels.

subplot

A two-element vector that sets out the number of rows and columns for a grid layout of multiple colorings.

clear

If TRUE, all graphics devices are cleared when the plot is refreshed.

interactive

If TRUE, an interactive version of the plot is launched.

file

If non-empty, the figure is saved as an SVG or HTML file instead of plotting on graphics device.

font

Multiplier to adjust font size for SVG and HTML output.

Details

The input topology must follow the format from nroKohonen(), but may also contain the columns REGION, and REGION.label that specify the names for subsets of districts and the single character labels to be shown on top of those districts. The input can also be the list object as returned by nroKohonen().

The color input can include the attribute 'contrast' that contains a binary vector or a matrix of equal size. If an element is set, it means a dark label or highlight will have better contrast with the background.

The label input can include the attribute 'visible' that contains a binary vector or a matrix of equal size. If an element is set, it means a label is visible, otherwise it will not be shown on the map.

Some non-alphanumeric characters are not supported and will be automatically converted to "_". Too long labels or column names will be truncated.

The default value for clear is TRUE to prevent multiple plot windows from accumulating within the RStudio. If you are running R from the terminal or using detached devices, setting clear to FALSE will retain the current window when refreshing.

If the file name ends with ".html", an interactive HTML document is produced, otherwise an SVG document is created. We recommend opening the HTML file with a web browser to select regions on large maps (i.e. when the R plot window becomes too slow to use). The HTML page allows you to assign subgroups and to save the results as tab-delimited text.

Examples

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

# Detect binary columns.
dataset <- nroPreprocess(dataset, method = "")

# Prepare training data.
trvars <- c("CHOL", "HDL2C", "TG", "CREAT", "uALB")
trdata <- scale.default(dataset[,trvars])

# K-means clustering.
km <- nroKmeans(data = trdata)

# Self-organizing map.
sm <- nroKohonen(seeds = km)
sm <- nroTrain(map = sm, data = trdata)

# Assign data points into districts.
matches <- nroMatch(centroids = sm, data = trdata)

# Select a subset of variables and detect binary data.
vars <- c("AGE", "MALE", "uALB", "CHOL", "DIAB_KIDNEY", "DECEASED")
selected <- nroPreprocess(dataset[,vars], method = "")

# Calculate district averages for seleted variables.
vars <- c("AGE", "MALE", "uALB", "CHOL", "DIAB_KIDNEY", "DECEASED")
planes <- nroAggregate(topology = sm, districts = matches, data = selected)

# Estimate statistics.
stats <- nroPermute(map = sm, districts = matches, data = selected)

# Set visuals.
colrs <- nroColorize(values = planes, amplitudes = stats)
labls <- nroLabel(topology = sm, values = planes)

# Add subgrouping information.
topo <- sm$topology
topo$REGION <- ""
topo$REGION[1:8] <- "Center"
topo$REGION[9:21] <- "Perimeter"

# Add subgroup labels.
topo$REGION.label <- ""
topo$REGION.label[1:8] <- "C"
topo$REGION.label[9:21] <- "P"

# Add subgroup colors.
topo$REGION.color <- ""
topo$REGION.color[1:8] <- "#00f00060"
topo$REGION.color[9:21] <- "#f000f060"

# Plot colorings on screen.
nroPlot(topology = topo, colors = colrs, labels = labls)

# Save colorings in file.
#fn <- "colorings.html"
#n <- nroPlot.save(file = fn, topology = topo,
#    colors = colrs, labels = labls)
#cat(n, " bytes saved in '", fn, "'\n", sep="")