valPlot: Validation Measure Plots

Description

valPlot displays a plot of the scores for each selected validation measure.

Usage

valPlot(x, measures = measureNames(x), legend = TRUE, 
	legendLoc = "topright", main = NULL, pch = NULL, type = "b", 
	ask = prod(par("mfcol")) < length(measures) && dev.interactive(), ...)

Arguments

An object of class "optCluster".

measures

Character vector of the names of the validation measures to plot. Any number of choices is allowed.

legend

If TRUE, provides a legend.

legendLoc

Character string specifying the location of the legend.

main

Character string specifying the title of graph.

pch

Specifies the plotting characters to use.

type

A character string specifying the type of plot.

ask

If TRUE, the user is prompted before each plot.

...

Additional plotting parameters.

Details

The the biological homogeneity index (BHI), biological stability index (BSI), Dunn index, and silhouette width measures should all be maximized. The average proportion of non-overlap (APN), average distance (AD), average distance between means (ADM), figure of merit (FOM), and connectivity measures should all be minimized.

References

Brock, G., Pihur, V., Datta, S. and Datta, S. (2008). clValid: An R Package for Cluster Validation. Journal of Statistical Software 25(4), http://www.jstatsoft.org/v25/i04.

Examples

Run this code

## This example may take a few minutes to compute
	## Obtain Dataset	
	data(arabid)	
	
	## Normalize Data with Respect to Library Size	
	obj <- t(t(arabid)/colSums(arabid))
		
	## Analysis of Normalized Data using Internal and Stability Validation Measures
	norm1 <- optCluster(obj, 2:4, clMethods = "all")
	
	## Plots of Internal and Stability Validation Measures
	par(mfrow = c(4,2))
	valPlot(norm1)
	
	## Plots of Internal Validation Measures in a Single Figure
	par(mfrow = c(2,2))
	valPlot(norm1, measure = c("Dunn", "Silhouette", "Connectivity"), legend = FALSE)
	plot(2:4, valScores(norm1, "Dunn")[,,1], type="n", axes=FALSE, xlab = "", ylab = "")
	legend("center", methodNames(norm1), col=1:9, lty=1:9, pch=paste(c(1:9)), 
			cex=0.8)