Re: [igraph] Re: community.to.membership

[Gábor Csárdi]

Hi,

I am not sure what kind of advice you are expecting.

If you have more than two connected components in your graph, then
obviously you cannot have two communities, at least it does not make
any sense to have two.

Gabor

On Tue, Nov 9, 2010 at 9:17 AM, Susan Biancani <address@hidden> wrote:
> I realized: I have isolates in my graph.  I assume this is the cause of the
> trouble, but am open to other advice/thoughts as well.
>
> On Mon, Nov 8, 2010 at 9:54 PM, Susan Biancani <address@hidden> wrote:
>>
>> Hi,
>> I'd like to identify two communities in my graph, using the
>> edge.betweenness.community algorithm.  But when I try to run
>> community.to.membership, I'm getting an error I don't understand.
>>
>> eb <- edge.betweenness.community(G)
>> pos.comm <- community.to.membership(G, eb$merges, steps=(length(V(G)))-2)
>>
>> > pos.comm <- community.to.membership(G, eb$merges,
>> > steps=(length(V(G)))-2)
>> Error in community.to.membership(G, eb$merges, steps = (length(V(G))) -  :
>>   At community.c:465 : `steps' to big or `merges' matrix too short,
>> Invalid value
>> >
>>
>> My graph has 500 nodes, and eb$merges has length of 808. I chose
>> steps=length(V(G))-2 in order to end up with two clusters in the end.  When
>> I try this on a smaller subset of this graph, containing 50 nodes, and a
>> merge matrix of length 98, this code works.
>>
>> Any suggestions would be appreciated!  Thanks,
>> Susan
>
>
> _______________________________________________
> igraph-help mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/igraph-help
>
>



-- 
Gabor Csardi <address@hidden>     UNIL DGM