[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [igraph] Help with using igraph_centralization_betweenness
From: |
Tamás Nepusz |
Subject: |
Re: [igraph] Help with using igraph_centralization_betweenness |
Date: |
Tue, 17 Jul 2012 12:08:29 +0200 |
Hi,
Thank you, I have managed to reproduce your issue and this is indeed a bug: one
of the variables overflow when igraph calculates the theoretical maximum of the
betweenness centrality in igraph_centralization_betweenness_tmax. The
workaround for the time being is as follows:
1. Call igraph_centralization_betweenness without normalization, keep the
centralization score but ignore the theoretical maximum.
2. Calculate the theoretical maximum as follows:
igraph_real_t vcount = igraph_vcount(graph);
igraph_real_t tmax = (vcount-1) * (vcount-1) * (vcount-2);
If (igraph_is_directed(graph))
tmax /= 2;
tmax then contains the theoretical maximum of the betweenness score.
3. Divide the centralization score with the theoretical maximum to get the
centralization score.
This will be fixed in the next minor release of igraph.
Best,--
T.
On Tuesday, 17 July 2012 at 11:13, Giuseppe G. wrote:
> Hi,
>
> my graph has 3695 nodes and 9814 edges.
>
> On 13/07/12 11:37, Tamás Nepusz wrote:
> > Hi,
> >
> > > Centralization - betweenness 4555078144.000000, max -980219904.000000
> > >
> > > Any other suggestions on what I might be doing wrong? I played with the
> > > other boolean parameters with no success unfortunately.
> >
> >
> >
> > How large your graph is? Centralization works fine for me the way you did
> > it on a small random graph -- see the attached C source.
> >
> > Best,
> > T.
>
>
>
>
>
> --
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
>
>
> _______________________________________________
> igraph-help mailing list
> address@hidden (mailto:address@hidden)
> https://lists.nongnu.org/mailman/listinfo/igraph-help