sna (version 0.3)

gclust.centralgraph: Get Central Graphs Associated with Graph Clusters

Description

Calculates central graphs associated with particular graph clusters (as indicated by the k partition of h).

Usage

gclust.centralgraph(h, k, mats, ...)

Arguments

h

An hclust object, the based on a graph set

k

The number of groups to evaluate

mats

A graph stack containing the adjacency matrices for the graphs on which the clustering was performed

...

Additional arguments to centralgraph

Value

• An array containing the stack of central graph adjacency matrices

Details

gclust.centralgraph uses cutree to cut the hierarchical clustering in h into k groups. centralgraph is then called on each cluster, and the results are returned as a graph stack. This is a useful tool for interpreting clusters of (labeled) graphs, with the resulting central graphs being subsequently analyzed using standard SNA methods.

References

Butts, C.T., and Carley, K.M. (2001). ``Multivariate Methods for Interstructural Analysis.'' CASOS working paper, Carnegie Mellon University.

See Also

hclust, centralgraph, gclust.boxstats, gdist.plotdiff, gdist.plotstats

Examples

#Create some random graphs
g<-rgraph(10,20,tprob=c(rbeta(10,15,2),rbeta(10,2,15)))

#Find the Hamming distances between them
g.h<-hdist(g)

#Cluster the graphs via their Hamming distances
library(mva)          #Load the mva library
g.c<-hclust(as.dist(g.h))

#Now find central graphs by cluster for a two cluster solution
g.cg<-gclust.centralgraph(g.c,2,g)

#Plot the central graphs
gplot(g.cg[1,,])
gplot(g.cg[2,,])