Re: [igraph] Possible bug transitivity for barabasi.game

[Fabio Daolio]

If I'm not mistaken,

you're generating a network by adding one vertex forming one link at a time;

in this case your network will have a "tree-like" appearance and it will not be

possible for you to have links forming triangular closures among neighbors.

Hence the null clustering coefficient.

You can try, with a smaller value of n, to plot the network to see what I mean.

In order to change this behavior, I think you should provide the generating 

a value greater than one to the parameter m, i.e. the number of links an

added vertex forms with the existing ones.

- please have a look at the function manual ?barabasi.game -

g1 <- barabasi.game(10000, m=1)

g2 <- barabasi.game(10000, m=2)

g3 <- barabasi.game(10000, m=3)

transitivity(g1)

transitivity(g2)

transitivity(g3)

On 19 oct. 2010, at 03:13, Anindya Ghosh wrote:

I am calculating the clustering coefficients of networks generated using the following code

n <- 10000

g <- barabasi.game(n)

transitivity(g)

I always get transitivity as 0 irrespective of the value of n. Is this is a bug? 

According to the literature it should not be 0. Any help will be appreciated.

--
"I'm not confused, I'm well mixed." -Robert Frost

_______________________________________________
igraph-help mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/igraph-help

--

Fabio