Re: [igraph] computing betweenness for weighted network
> Are there any standards for the edge and vertex attributes? For > example, can I assume that for a simple weighted graph the edge weight > will always be called "weight"? Or are there no guarantees? Weights are always stored in "weight" (edge attribute) and vertex names are always stored in "name" - this is because igraph's is_named() and is_weighted() methods also look for these attributes. Also, for bipartite graphs, the two parts of the graph are encoded in the "type" vertex attribute. T. ___ igraph-help mailing list igraph-help@nongnu.org https://lists.nongnu.org/mailman/listinfo/igraph-help
Re: [igraph] computing betweenness for weighted network
On 4 May 2016 at 15:59, Tamas Nepusz wrote: >> I didn't know about Nexus. Is it still being maintained? > Well -- the server is operational, the API has not changed, so it > works, and we would probably fix it if someone reports that it is not > functioning properly. However, we haven't received any data donations > and probably lack the resources (bandwidth and storage space) to host > large-scale datasets on our servers. Are there any standards for the edge and vertex attributes? For example, can I assume that for a simple weighted graph the edge weight will always be called "weight"? Or are there no guarantees? ___ igraph-help mailing list igraph-help@nongnu.org https://lists.nongnu.org/mailman/listinfo/igraph-help
Re: [igraph] computing betweenness for weighted network
> I didn't know about Nexus. Is it still being maintained? Well -- the server is operational, the API has not changed, so it works, and we would probably fix it if someone reports that it is not functioning properly. However, we haven't received any data donations and probably lack the resources (bandwidth and storage space) to host large-scale datasets on our servers. T. ___ igraph-help mailing list igraph-help@nongnu.org https://lists.nongnu.org/mailman/listinfo/igraph-help
Re: [igraph] computing betweenness for weighted network
I didn't know about Nexus. Is it still being maintained?
On 4 May 2016 at 15:25, Tamas Nepusz wrote:
> Hi,
>
>> I would like to compute the betweenness centrality of the weighted zachary
>> karate club.
>> [...]
>> I would like to have the same result as in this paper
>
> You did not mention how did you obtain the Zachary karate club
> network. If you simply ran this:
>
> g = Graph.Famous("zachary")
>
> then it's no good because this generates the _unweighted_ Zachary
> karate club network.
>
> However, you could download the weighted version of the network from
> the Nexus graph repository (http://nexus.igraph.org) with the
> following command:
>
> g = Nexus.get("karate")
> g.betweeness(weights="weight")
>
> Then you would get slightly different results that are somewhat closer
> to the results reported in that paper. They are still not the same,
> but I cannot know whether it's due to the fact that the version of the
> network in Nexus is different from the network that the authors have
> used in that paper, or because something is wrong either in our
> calculations or theirs.
>
> T.
>
> ___
> igraph-help mailing list
> igraph-help@nongnu.org
> https://lists.nongnu.org/mailman/listinfo/igraph-help
___
igraph-help mailing list
igraph-help@nongnu.org
https://lists.nongnu.org/mailman/listinfo/igraph-help
Re: [igraph] computing betweenness for weighted network
Hi,
> I would like to compute the betweenness centrality of the weighted zachary
> karate club.
> [...]
> I would like to have the same result as in this paper
You did not mention how did you obtain the Zachary karate club
network. If you simply ran this:
g = Graph.Famous("zachary")
then it's no good because this generates the _unweighted_ Zachary
karate club network.
However, you could download the weighted version of the network from
the Nexus graph repository (http://nexus.igraph.org) with the
following command:
g = Nexus.get("karate")
g.betweeness(weights="weight")
Then you would get slightly different results that are somewhat closer
to the results reported in that paper. They are still not the same,
but I cannot know whether it's due to the fact that the version of the
network in Nexus is different from the network that the authors have
used in that paper, or because something is wrong either in our
calculations or theirs.
T.
___
igraph-help mailing list
igraph-help@nongnu.org
https://lists.nongnu.org/mailman/listinfo/igraph-help
[igraph] computing betweenness for weighted network
HI all I would like to compute the betweenness centrality of the weighted zachary karate club. That network has weights. What command should I used? I would like to have the same result as in this paper https://www.semanticscholar.org/paper/Laplacian-centrality-A-new-centrality-measure-for-Qi-Fuller/d1939e822b2bc828c13734f9240485aec25a1ded Cheers I issued this command karate.betweenness () and i got [231.0714285714286, 28.478571428571424, 75.85079365079366, 6.288095238095237, 0., 15.834, 15.832, 0.0, 29.529365079365082, 0.44761904761904764, 0., 0.0, 0.0, 24.215873015873015, 0.0, 0.0, 0.0, 0.0, 0.0, 17.1468253968254, 0.0, 0.0, 0.0, 9.299, 1.1665, 2.0277, 0.0, 11.79206349206349, 0.9476190476190476, 1.5428571428571427, 7.60952380952381, 73.00952380952378, 76.69047619047619, 160.5515873015873] but those values seem no to be the same as in the paper here https://www.semanticscholar.org/paper/Laplacian-centrality-A-new-centrality-measure-for-Qi-Fuller/d1939e822b2bc828c13734f9240485aec25a1ded cheers ___ igraph-help mailing list igraph-help@nongnu.org https://lists.nongnu.org/mailman/listinfo/igraph-help
