Question

I’m using the Walktrap community detection method to return a number (19 in this case) of clusters. I have a list of members which belong to one or more of these clusters.

1. I need a method to search each cluster for the presence of the
   members and return the percentage of matches found. ( e.g cluster[0]
   = 0%, cluster[1] =Y%…..cluster[18]=Z%) Thus selecting the optimum cluster that represents the members on the list.

2. Once the optimum cluster is found, I need a method to count the
   number of members of the optimum cluster and from the original
   (19-1) clusters select another cluster that is nearest in size
   (number of members)

    library(igraph)
    edges <- read.csv('http://dl.dropbox.com/u/23776534/Facebook%20%5BEdges%5D.csv')
    list<-read.csv("http://dl.dropbox.com/u/23776534/knownlist.csv")
    all<-graph.data.frame(edges)
    summary(all)
   all_wt<- walktrap.community(all, steps=6,modularity=TRUE,labels=TRUE)
   all_wt_memb <- community.to.membership(all,all_wt$merges,steps=which.max(all_wt$modularity)-1)
   all_wt_memb$csize
   
   >[1] 176  13 204  24   9 263  16   2   8   4  12   8   9  19  15   3   6   2   1
   

Answer 1

The %in% function, when used like: a %in% b will determine which of the elements in vector a are also present in vector b. So for each cluster, I would

• Extract the members of that cluster
• Given a list of members in which you’re interested, calculate which ones are %in% this cluster — which will return a Boolean vector
• You can use sum() on the Boolean vector to count the number of true elements (i.e. the number of elements in your initial vector which are present in this cluster
• (Optionally) you can normalize by the length of the cluster to get the percentage of this cluster which is made up of your list of interest, or by the length of the list you made, to indicate the number of members in your list which are present in this cluster.

You can loop through each cluster using for() or an apply variant.

Then given all_wt_memb$csize, you’ll have a given value which is your target, and you’ll want to find the nearest number. See this link, but you’re just calculating the minimum absolute difference:

x=c(1:100)
your.number=5.43
which(abs(x-your.number)==min(abs(x-your.number)))