Re: [igraph] Community similarity / dissimilarity

[Tim Wilson]

Thank you Tamas

> Would you like to compare two communities, or to compare two community 
> structures as a whole?

vc = g.community_infomap()
n = vc.n

I'm trying to make an n by n similarity matrix that compares every
discovered community w/ every other discovered community

My ultimate goal is to use this similarity matrix and multidimensional
scaling to make a nice 2d layout of the discovered communities

> If you wish to compare two given communities only, are they from the same 
> clustering of the same graph (i.e. are they disjoint) or are they from 
> different clusterings of the same graph?

They're disjoint



Tim



On Wed, Dec 19, 2012 at 3:34 PM, Tamás Nepusz <address@hidden> wrote:
> Hi,
>
>> I'm trying to identify a good distance (similarity / dissimilarity)
>> measure for discovered VertexDendrogram and VertexClustering
>> communities
> Would you like to compare two communities, or to compare two community 
> structures as a whole? For the latter, we have a function called 
> compare_communities, which implements several distance/similarity metrics on 
> VertexClustering objects.
>
> If you wish to compare two given communities only, are they from the same 
> clustering of the same graph (i.e. are they disjoint) or are they from 
> different clusterings of the same graph?
>
> Best,
> Tamas
>
>
> _______________________________________________
> igraph-help mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/igraph-help