How about contracting the graph first, and then removing the vertices
that belong to the small communities after?
G.
On Sun, Nov 24, 2013 at 4:34 PM, <address@hidden> wrote:
I have an object "communities" with few large communities and many
communities with just few members. I would like to create a community
graph
to plot (with one node for each community) but excluding communities
with
less than 20 members.
I was able to subset the graph based on membership to largest
communities
with
ids <- which(sizes(communities)>20)
subgraph <- induced.subgraph(graph, which(membership(communities)
%in%
id))
Now I wanted to apply this (from
http://stackoverflow.com/a/12699979/1707938) to my case
contractgraph <- contract.vertices(subgraph,
membership(communities))
E(contractgraph)$weight <- 1
contractgraph2 <- simplify(contractgraph, remove.loops=FALSE)
But of course I have a problem because "subgraph" and "communities"
have
different length. How could I proceed? Should I try to subset
"communities"
based on sizes(communities)>20?
_______________________________________________
igraph-help mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/igraph-help
_______________________________________________
igraph-help mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/igraph-help