Re: [igraph] community detection algorithm and Evaluation

[Tamas Nepusz]

as I have two types of graph one is directed an weighted and the other one is undirected and unweighted, the one which I could use for both are four (1,2,4,5) which I get the error on the forth one as my graph is an unconnected graph, so there is three.

You could still use the fourth one by decomposing your graph into connected components first (see ?decompose.graph), calculating the communities for each of the components, and then merging the community membership vectors.

• is there any other algorithm which is implemented in igraph and is not in the list? and which will give me overlapping communities as well.

No, but clique percolation is not particularly hard to implement in igraph - the naive solution would work for graphs of moderate size:

http://igraph.wikidot.com/community-detection-in-r#toc0

• 
  which of these metric could be used for weighted and directed graph and is there any implementation in igraph?

modularity works for weighted graphs but ignores edge directions (since there is no agreement on the scientific community yet about how to extend modularity for directed networks; several competing proposals have been described in the literature). compare.communities() does not care about the graph since it compares the communities with a ground truth, so it does not matter whether the graph was directed or not.

• 
  also which metric could be used for which algorithm? , as I go through one of the article "edge-betweeness"the metric used in there was the ground truth and they compare to the known community graph.

You could use any of the metrics with any of the algorithms. Keep in mind that some of the algorithms explicitly try to optimise the modularity behind the scenes (one way or another).

T.