hello Tamas,

sorry I didn't have my computer available until now. I can send my
data for sure, but before I waste your time, I see that cluster(g,
mode="strong") gives me 27 clusters where 26 are made of 1 node and
one is made of 30 nodes. Then table(graph.coreness(g, mode="in"))
gives me:

 0  6  7  8 10
 9  3  2  7 35

whille table(graph.coreness(g, mode="out")) gives:

 0  4  5  6  7  8  9
17  2  2  2  2  9 22

are these good indicators that maybe the network has a core-periphery
structure, with nested k-groups, instead of connected communities with
some links between them? If yes, then I guess
edge.betweenness.community is right in saying that my network has just
one community...

again, if this is a suitable answer, is there any core-periphery
algorithm in igraph?

thanks for your assistance,
Simone


Sent from my iPhone


Il giorno 21/mag/2012, alle ore 14:52, Tamás Nepusz <[email protected]>
ha scritto:

>> I have:
>> Graph community structure calculated with the edge betweenness algorithm
>> Number of communities (best split): 1
>> Modularity (best split): 0
>>
>> with igraph 0.6, and a network of 56 nodes and 813 edges... should I
>> fine-tune something? looks like just one community exists...
>
> Could be two things:
>
> 1) it is indeed the case that only one community exists (at least according 
> to the algorithm)
> 2) there is a bug in the weighted variant of the edge betweenness community 
> detection, which has just been added recently to 0.6.
>
> If you send me your graph and the weights, I can try to distinguish between 
> the two.
>
> Best,
> T.
>
>
> _______________________________________________
> igraph-help mailing list
> [email protected]
> https://lists.nongnu.org/mailman/listinfo/igraph-help

_______________________________________________
igraph-help mailing list
[email protected]
https://lists.nongnu.org/mailman/listinfo/igraph-help

Reply via email to