Re: [igraph] triangle related question

[Gábor Csárdi]

Hi Simone,

well, you can plot the mean local transitivity in the function of
degree easily. I think this is better than the number of triangles,
because that is biased in the sense that even in a Gnp random network
the degree and the number of triangles are correlated, simply because
high degree vertices have more possibilities to participate in
triangles.

The code is something like

g <- forest.fire.game(10000, fw.prob=0.37, bw.factor=0.32/0.37)
tra <- transitivity(g, type="local")
tra.deg <- tapply(tra, degree(g), mean)
deg <- as.numeric(names(tra.deg))
plot(deg, tra.deg, log="xy")        # for this example log-log looks better

Best,
G.

On Sat, Mar 28, 2009 at 3:38 PM, simone gabbriellini
<address@hidden> wrote:
> hi list,
>
> I send again this message because the first one was in html format. sorry
> for the cross-post.
>
> can you help me to plot the degree d of node vs. the mean number of
> triangles over all nodes with degree d?
>
> I am trying to see if my networks exibit the same relations as in
> Tsourakakis "Fast Counting of Triangles in Large Real Networks: Algorithms
> and Laws", figure 7.
>
> many thanks,
> Simone
>
>
>
>
>
> _______________________________________________
> igraph-help mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/igraph-help
>



-- 
Gabor Csardi <address@hidden>     UNIL DGM