Learn R Programming

psych (version 1.0-17)

ICLUST.graph: create control code for ICLUST graphical output

Description

Given a cluster structure determined by ICLUST, create dot code to describe the ICLUST output. To use the dot code, use either http://www.graphviz.org/ Graphviz or a commercial viewer (e.g., OmniGraffle).

Usage

ICLUST.graph(ic.results, out.file,min.size=1, short = FALSE,labels=NULL,
size = c(8, 6), node.font = c("Helvetica", 14), edge.font = c("Helvetica", 12), 
rank.direction = "RL", digits = 2, title = "ICLUST", ...)

Arguments

ic.results
output list from ICLUST
out.file
name of output file (defaults to console)
min.size
draw a smaller node (without all the information) for clusters < min.size -- useful for large problems
short
if short==TRUE, don't use variable names
labels
vector of text labels (contents) for the variables
size
size of output
node.font
Font to use for nodes in the graph
edge.font
Font to use for the labels of the arrows (edges)
rank.direction
LR or RL
digits
number of digits to show
title
any title
...
other options to pass

Value

  • Output is a set of dot commands written either to console or to the output file. These commands may then be used as input to any "dot" viewer, e.g., Graphviz.

Details

Will create (or overwrite) an output file and print out the dot code to show a cluster structure. This dot file may be imported directly into a dot viewer (e.g., http://www.graphviz.org/). The "dot" language is a powerful graphic description language that is particulary appropriate for viewing cluster output. Commercial graphics programs (e.g., OmniGraffle) can also read (and clean up) dot files. ICLUST.graph takes the output from ICLUST results and processes it to provide a pretty picture of the results. Original variables shown as rectangles and ordered on the left hand side (if rank direction is RL) of the graph. Clusters are drawn as ellipses and include the alpha, beta, and size of the cluster. Edges show the cluster intercorrelations.

It is possible to trim the output to not show all cluster information. Clusters < min.size are shown as small ovals without alpha, beta, and size information.

References

ICLUST: http://personality-project.org/r/r.iclust.html

See Also

VSS.plot, ICLUST

Examples

Run this code
test.data <- Harman74.cor$cov
ic.out <- ICLUST(test.data)
out.file <- file.choose(new=TRUE)   #create a new file to write the plot commands to 
ICLUST.graph(ic.out,out.file)   
now go to graphviz (outside of R) and open the out.file you created
print(ic.out,digits=2)

 
#test.data <- Harman74.cor$cov 
#my.iclust <- ICLUST(test.data)
#ICLUST.graph(my.iclust)
#
#
#digraph ICLUST {
#  rankdir=RL;
#  size="8,8";
#  node [fontname="Helvetica" fontsize=14 shape=box, width=2];
#  edge [fontname="Helvetica" fontsize=12];
# label = "ICLUST";
#	fontsize=20;
#V1  [label = VisualPerception];
#V2  [label = Cubes];
#V3  [label = PaperFormBoard];
#V4  [label = Flags];
#V5  [label = GeneralInformation];
#V6  [label = PargraphComprehension];
#V7  [label = SentenceCompletion];
#V8  [label = WordClassification];
#V9  [label = WordMeaning];
#V10  [label = Addition];
#V11  [label = Code];
#V12  [label = CountingDots];
#V13  [label = StraightCurvedCapitals];
#V14  [label = WordRecognition];
#V15  [label = NumberRecognition];
#V16  [label = FigureRecognition];
#V17  [label = ObjectNumber];
#V18  [label = NumberFigure];
#V19  [label = FigureWord];
#V20  [label = Deduction];
#V21  [label = NumericalPuzzles];
#V22  [label = ProblemReasoning];
#V23  [label = SeriesCompletion];
#V24  [label = ArithmeticProblems];
#node [shape=ellipse, width ="1"];
#C1-> V9 [ label = 0.78 ];
#C1-> V5 [ label = 0.78 ];
#C2-> V12 [ label = 0.66 ];
#C2-> V10 [ label = 0.66 ];
#C3-> V18 [ label = 0.53 ];
#C3-> V17 [ label = 0.53 ];
#C4-> V23 [ label = 0.59 ];
#C4-> V20 [ label = 0.59 ];
#C5-> V13 [ label = 0.61 ];
#C5-> V11 [ label = 0.61 ];
#C6-> V7 [ label = 0.78 ];
#C6-> V6 [ label = 0.78 ];
#C7-> V4 [ label = 0.55 ];
#C7-> V1 [ label = 0.55 ];
#C8-> V16 [ label = 0.5 ];
#C8-> V14 [ label = 0.49 ];
#C9-> C1 [ label = 0.86 ];
#C9-> C6 [ label = 0.86 ];
#C10-> C4 [ label = 0.71 ];
#C10-> V22 [ label = 0.62 ];
#C11-> V21 [ label = 0.56 ];
#C11-> V24 [ label = 0.58 ];
#C12-> C10 [ label = 0.76 ];
#C12-> C11 [ label = 0.67 ];
#C13-> C8 [ label = 0.61 ];
#C13-> V15 [ label = 0.49 ];
#C14-> C2 [ label = 0.74 ];
#C14-> C5 [ label = 0.72 ];
#C15-> V3 [ label = 0.48 ];
#C15-> C7 [ label = 0.65 ];
#C16-> V19 [ label = 0.48 ];
#C16-> C3 [ label = 0.64 ];
#C17-> V8 [ label = 0.62 ];
#C17-> C12 [ label = 0.8 ];
#C18-> C17 [ label = 0.82 ];
#C18-> C15 [ label = 0.68 ];
#C19-> C16 [ label = 0.66 ];
#C19-> C13 [ label = 0.65 ];
#C20-> C19 [ label = 0.72 ];
#C20-> C18 [ label = 0.83 ];
#C21-> C20 [ label = 0.87 ];
#C21-> C9 [ label = 0.76 ];
#C22-> 0 [ label = 0 ];
#C22-> 0 [ label = 0 ];
#C23-> 0 [ label = 0 ];
#C23-> 0 [ label = 0 ];
#C1  [label =   "C1\n  alpha= 0.84\n beta=  0.84\nN= 2"] ;
#C2  [label =   "C2\n  alpha= 0.74\n beta=  0.74\nN= 2"] ;
#C3  [label =   "C3\n  alpha= 0.62\n beta=  0.62\nN= 2"] ;
#C4  [label =   "C4\n  alpha= 0.67\n beta=  0.67\nN= 2"] ;
#C5  [label =   "C5\n  alpha= 0.7\n beta=  0.7\nN= 2"] ;
#C6  [label =   "C6\n  alpha= 0.84\n beta=  0.84\nN= 2"] ;
#C7  [label =   "C7\n  alpha= 0.64\n beta=  0.64\nN= 2"] ;
#C8  [label =   "C8\n  alpha= 0.58\n beta=  0.58\nN= 2"] ;
#C9  [label =   "C9\n  alpha= 0.9\n beta=  0.87\nN= 4"] ;
#C10  [label =   "C10\n  alpha= 0.74\n beta=  0.71\nN= 3"] ;
#C11  [label =   "C11\n  alpha= 0.62\n beta=  0.62\nN= 2"] ;
#C12  [label =   "C12\n  alpha= 0.79\n beta=  0.74\nN= 5"] ;
#C13  [label =   "C13\n  alpha= 0.64\n beta=  0.59\nN= 3"] ;
#C14  [label =   "C14\n  alpha= 0.79\n beta=  0.74\nN= 4"] ;
#C15  [label =   "C15\n  alpha= 0.66\n beta=  0.58\nN= 3"] ;
#C16  [label =   "C16\n  alpha= 0.65\n beta=  0.57\nN= 3"] ;
#C17  [label =   "C17\n  alpha= 0.81\n beta=  0.71\nN= 6"] ;
#C18  [label =   "C18\n  alpha= 0.84\n beta=  0.75\nN= 9"] ;
#C19  [label =   "C19\n  alpha= 0.74\n beta=  0.65\nN= 6"] ;
#C20  [label =   "C20\n  alpha= 0.87\n beta=  0.74\nN= 15"] ;
#C21  [label =   "C21\n  alpha= 0.9\n beta=  0.77\nN= 19"] ;
#C22  [label =   "C22\n  alpha= 0\n beta=  0\nN= 0"] ;
#C23  [label =   "C23\n  alpha= 0\n beta=  0\nN= 0"] ;
#{ rank=same;
#V1;V2;V3;V4;V5;V6;V7;V8;V9;V10;V11;V12;V13;V14;V15;V16;V17;V18;V19;V20;V21;V22;V23;V24;}}
#
#copy the above output to Graphviz and draw it
#see \url{http://personality-project.org/r/r.ICLUST.html} for an example.

## The function is currently defined as
function(ic.results, out.file,short=FALSE,
   size=c(8,8), node.font=c("Helvetica", 14),
    edge.font=c("Helvetica", 12), rank.direction="RL", digits=2,title="ICLUST", ...){
    
      if(!missing(out.file)){
        out <- file(out.file, "w")
        on.exit(close(out))
        }
        else out <- stdout()
    results <- ic.results$results
    var.labels <- rownames(ic.results$loadings)
    clusters <- ic.results$clusters
    if(length(clusters)==length(var.labels) ){clusters <- as.matrix(clusters)}
	num <- nrow(results)
    if (short) {var.labels <- paste("V",1:nrow(var.labels),sep="")}
   
  rank.direction <- match.arg(rank.direction)
  #first some basic setup parameters 
  cat( file=out,paste('digraph ICLUST', ' {<n>', sep=""))
  cat(file=out, paste('  rankdir=', rank.direction, ';<n>', sep=""))
  cat(file=out, paste('  size="',size[1],',',size[2],'";<n>', sep=""))
  cat(file=out, paste('  node [fontname="', node.font[1], 
        '" fontsize=', node.font[2], ' shape=box, width=2];<n>', sep=""))
  cat(file=out, paste('  edge [fontname="', edge.font[1],
        '" fontsize=', edge.font[2], '];<n>', sep=""))
  cat(file=out, paste(' label = "' ,title,'";
	fontsize=20;<n>', sep=""))
 
   #create the items as boxes  
   #add the sign from the clusters 
  num.var <- nrow(results)+1   #how many variables?
     for (i in 1:num.var) { if (max(clusters[i,]) > 0 ) {
      	cat(file=out,paste('V',i,'  [label = "',var.labels[i], '"];<n>', sep=""))</n> else {
      	cat(file=out,paste('V',i,'  [label = "-',var.labels[i], '"];<n>', sep=""))</n>
     	 }
   
  #show the cluster structure with ellipses
  
  cat(file=out,paste('node [shape=ellipse, width ="1"];<n>', sep=""))
  for (i in 1:num) {if(results[i,1]>0) { #avoid printing null results
     cat(file=out,paste(row.names(results)[i],  '-> ', results[i,1],
     ' [ label = ',round(results[i,"r1"],digits),' ];<n>', sep=""))
     cat(file=out,paste(row.names(results)[i],  '-> ', results[i,2], 
     ' [ label = ',round(results[i,"r2"],digits),' ];<n>', sep=""))</n></n>
    
   #label the clusters with alpha and beta
   for (i in 1:num) {if(results[i,1]>0) { #don't print blank results
     cat(file=out,paste(row.names(results)[i],  '  [label =   "',row.names(results)[i],
     '\n  alpha= ',round(results[i,"alpha"],digits),'\n beta=  ' ,
     round(results[i,"beta"],digits),'\nN= ',results[i,"size"], '"] ;<n>', sep=""))</n>}
  
  #keep the boxes all at the same rank (presumably the left side)
  cat(file=out, paste('{ rank=same;<n>', sep=""))
  for (i in 1:num.var) { cat(file=out,paste('V',i,';', sep=""))
  }   
   cat(file=out, paste('</n>}', sep=""))   # we are finished
 } # end of ICLUST.graph 
}

<keyword>multivariate</keyword><keyword>cluster</keyword></n></n></n></n></n></n></n>

Run the code above in your browser using DataLab