Re: [igraph] Re: [statnet_help] Re: gplot.layout with valued edges

[Gábor Csárdi]

On Tue, Mar 17, 2009 at 3:50 PM, Alexander <address@hidden> wrote:
> Hi,
>
> thanks for the suggestion about the new version. I installed it and
> I'm now trying to replicate some Fowler's results to test it.
>
> I have another question:
>
> I want to examine the fact that the nodes in my network are divided
> into two groups based on an attibute. This is why I was using the
> brokerage measure of the sna package. However, I want a measure that
> takes into account the weight of the edges. So can I use closeness()
> but with the within / between group distinction in some way?

There is no easy way to do this I think. What you can try is
calculating the weighted shortest paths for all pairs, but then drop
the ones you are not interested in and calculate the closeness by
hand.

Best,
Gabor

> looking at the r interface there does not seem to be a way to define
> the 'to' vertex set, but could I do this if I create two new weight
> attributes - one for within-group and one for between group, and then
> used those as weights?
>
> Again, thanks
>
> Alexander
>
> On 3/17/09, Jose Quesada <address@hidden> wrote:
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> Gábor Csárdi wrote:
>>> Alexander,
>>>
>>> On Tue, Mar 17, 2009 at 1:07 AM, Alexander Jerneck
>>> <address@hidden> wrote: [...]
>>>> Newman's measure of betweeness centrality for scientific
>>>> collaborations [1]
>>>
>>> The (not yet finished) 0.6 version of igraph has weighted
>>> betweenness, I think this is the same as the measure in Newman's
>>> paper. You just need to calculate the weights of the edges and call
>>> betweenness().
>>>
>>> You can download an R source package from here:
>>> http://cneurocvs.rmki.kfki.hu/igraph/download/igraph_0.6.tar.gz and
>>> an R package for windows from here:
>>> http://cneurocvs.rmki.kfki.hu/igraph/download/igraph_0.6.zip
>>>
>>>> Fowler's adaptation of this measure to directed networks (of
>>>> cosponsorship in the US Congress) [2]
>>>
>>> Version 0.6 also has weighted closeness, and this looks the same to
>>> me as Fowler's measure of 'connectedness'. Again, just calculate
>>> the weights, and then call closeness().
>>>
>>> Best, Gabor
>>>
>>> [...]
>>>
>> Just a quick note, the windows package there is pretty old, and
>> there's one typo on graph.incidence().
>> It will return:
>> Error in graph.incidence.dense(indicende, directed = directed, mode =
>> mode,  :
>>   object "weigted" not found
>>
>> This has been fixed in more recent 0.6 versions.
>>
>> Best
>> - -Jose
>>
>> - --
>> Jose Quesada, PhD.
>> Max Planck Institute,
>> Center for Adaptive Behavior and cognition,
>> Berlin
>> http://www.josequesada.name/
>> -----BEGIN PGP SIGNATURE-----
>> Version: GnuPG v1.4.7 (MingW32)
>> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
>>
>> iQIVAwUBSb+WwmMobUVGH+HKAQJKMQ/8CUw5o5D4+tSjypzLGYcewjWaXoDmulck
>> Zm6bya5EKGVommRRgvQs6XRYvL5H7r9OmUo0IlVqp40WgwKxPRIyx91jqew+tPLn
>> bh7v/ExotYIztF3qxd2DBmCOmV41puZd27UWb6oMZ+zp97hUakRRAmjNuDVNza4s
>> 8cArRT8AXwJ+5nqP+Kj1yvQINgoOXF/TQuOYZrTUG3kBNthVb9EdYTffyl5zIVdq
>> ASqXj87mfJxwTa1qcndmv2h1JEAQODZ5yR8i1os4Ju8UiFsIxOiO+cY3oOHP89Ot
>> Xk4uCMZ2I0fobhWvAIP7Y9KuB129rKq1D0iRNdjWLu2SYPbC4o3acmc28VKx3VfN
>> v7B9LWydT9wrFYZbf6XigxEnY1HXp+B65IUbqx4f/Bx5X5cNFOjIxvGBoBa9ncaw
>> 4KnuKOeAmDhGJ2t4EXedRfPCw0ckwgBqBAalSHMRgtJVdgTPoBDEp3X5DtHM0oop
>> PRatGYGQB+7HO6+0N3w3x/ynF6qCuH2hJeUYuGqcpLUPFVZ/vfXwA4DmggZq6+/O
>> WBJHhWBtyyKnN5WP5lXJzyUIGYtL6k9ZA45HViXwrwko68VJTphmpIpnZgiFvyrz
>> QSq3sj5G0+oRMTyM1AYU7/B0WVVl18HWRzn84px3UOKuiq1B7aaJoEP3vfCJmXdh
>> W+80Pwr0tW8=
>> =IZzO
>> -----END PGP SIGNATURE-----
>>
>>
>>
>> _______________________________________________
>> igraph-help mailing list
>> address@hidden
>> http://lists.nongnu.org/mailman/listinfo/igraph-help
>>
>
>
> _______________________________________________
> igraph-help mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/igraph-help
>



-- 
Gabor Csardi <address@hidden>     UNIL DGM