Re: [igraph] how to generate directed small world networks

[simone gabbriellini]

Gabor,

Thanks again... I see the forest.fire is able to model the outdegree  
dist too... but the reciprocity is still zero.

I think I need to develop my own model. maybe mixing existing stuff.

The fact is that I need old vertex not only to be choosen by new ones,  
but also to act and to choose new ones as a response. I think that  
this mechanism would be sufficient to model my data.

barabasi.game is a function that I can modify? or does it call come C  
function I can't see?

Best,
Simone

Il giorno 15/mar/09, alle ore 11:34, Gábor Csárdi ha scritto:

> Simone,
>
> with barabasi.game you only model the in-degree distribution. What you
> can do is "plugging in" the out-degree distribution. But I think this
> is all, really. The reciprocity will be still zero.
>
> If you want to model both the in- and out-degree distribution, then
> you need another model. E.g. you can take a look at
> 'forest.fire.game'. Perhaps there are others in the literature.
>
> Best,
> Gabor
>
> On Sun, Mar 15, 2009 at 11:23 AM, simone gabbriellini
> <address@hidden> wrote:
> > Gabor,
> > yes, I mean that. Thank you for the suggestion. May I ask a  
> > question about
> > barabasi.game too?
> > I see that in each step it add a vertex and some edges to old  
> > vertices
> > already in the network with a certain probability. And that it is  
> > possible
> > to make attractive also the isolated vertex with the zero.appeal  
> > parameter.
> > g.emp is my empiric network. The indegree distribution produced with:
> > g.sf<-barabasi.game(n=g.emp.vcount, power=0.08, m=14, zero.appeal =  
> > 1.6)
> > is close to my data, as you can see
> > here: http://www.digitaldust.it/papers/analysis.pdf
> > (red = empiric, blue = scale free)
> > but the outdegree distribution is far from the empiric one...
> > I see also that the reciprocity(g.sf) is equal to 0, which is  
> > another bad
> > result for me...
> > do you have any suggestion in order to modify it to "improve" the  
> > fit of the
> > outdegree distribution?
> > or to increase the reciprocity?
> > I apologize if my questions are obvious or implicit in the nature  
> > of the
> > scale-free model.
> > Best,
> > Simone
> > Il giorno 15/mar/09, alle ore 11:09, Gábor Csárdi ha scritto:
> >
> > Simone, you mean Watts-Strogatz type? Create a lattice first, and  
> > then
> > call rewire. edges on it. E.g.
> >
> > g2 <- rewire.edges(graph.lattice(100, nei=2, dir=TRUE, mutual=TRUE),
> > p=2/100)
> > plot(g2, layout=layout.circle, vertex.label=NA, vertex.size=3,
> > edge.arrow.size=0.4)
> >
> > Best,
> > Gabor
> >
> > On Sat, Mar 14, 2009 at 3:17 PM, Simone Gabbriellini
> > <address@hidden> wrote:
> >
> > Dear List,
> >
> > is it possible to generate directed small world in the igraph R  
> > package?
> >
> > thanks,
> >
> > simone
> >
> >
> > _______________________________________________
> >
> > igraph-help mailing list
> >
> > address@hidden
> >
> > http://lists.nongnu.org/mailman/listinfo/igraph-help
> >
> >
> >
> >
> > --
> > Gabor Csardi <address@hidden>     UNIL DGM
> >
> >
> > _______________________________________________
> > 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
>
>
> _______________________________________________
> igraph-help mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/igraph-help