Re: [igraph] betweenness centrality -- negative and floating values?

[Tamas Nepusz]

> However, for my other network data, BC seems very easily to overflow.
> For instance, in the case of 4000 nodes which is not a so large
> network. I suggest that you should find some other solutions (probably
> the final) for this problem overflow/negative values, etc.
Yes. you're right, this is meant to be a temporary solution only (that's
why I didn't close the bug report yet). The proper solution would be
either to use a bignum library or avoid counting the geodesics
explicitly (since we only need a ratio of two geodesic counts in the
end). We will have to think about it, though. 

Best,
-- 
Tamas

pgpZH5mHS_fAS.pgp
Description: PGP signature