Re: [igraph] alpha centrality and bonpow fail with large graph

[Gábor Csárdi]

Michael,

the version you cited from the mailing list uses sparse matrices,
which will definitely help if your graphs are sparse.

It is possible to use sparse matrices for bonpow as well, see e.g.
http://igraph.wikidot.com/r-recipes#toc6

Nevertheless, R/igraph should not crash, so i'll take a look. Thanks
for the report!

Best,
Gabor

On Tue, May 3, 2011 at 5:28 PM, Michael Bishop <address@hidden> wrote:
> Dear Graphxperts,
>
> I have a graph g with Vertices: 67615  Edges: 251758 Directed: TRUE
>
> alpha centrality throws and error, i.e.
>
>> t <- data.frame(V(g)$name, alpha.centrality(g, alpha=.09))
> Error in matrix(0, nr = vcount(graph), nc = vcount(graph)) :
>  too many elements specified
>
> bonpow crashes R
>> t <- data.frame(V(g)$name, bonpow(g, exponent=.09))
>
> On 5/20/2008 Gabor said:
>  Hmmmm, 'alpha.centrality' is not very good, it uses the adjacency
>  matrix, it is thus inappropriate for large graphs. Here is what
>  you can do using the 'Matrix' package:
>
>  http://lists.gnu.org/archive/html/igraph-help/2008-04/msg00049.html
>
>  I hope it helps,
>  Gabor
>
> Is this suggestion from Gabor still the best option for alpha
> centrality?  What about bonpow?
>
> For what its worth, I've got 8gb of ram and would happily buy another
> 8 if it simplifies my work.
>
>> sessionInfo()
> R version 2.13.0 (2011-04-13)
> Platform: x86_64-pc-mingw32/x64 (64-bit)
>
> locale:
> [1] LC_COLLATE=English_United States.1252
> [2] LC_CTYPE=English_United States.1252
> [3] LC_MONETARY=English_United States.1252
> [4] LC_NUMERIC=C
> [5] LC_TIME=English_United States.1252
>
> attached base packages:
> [1] stats     graphics  grDevices utils
> [5] datasets  grid      methods   base
>
> other attached packages:
> [1] igraph_0.5.5-2 reshape_0.8.4  plyr_1.5.2
> [4] proto_0.3-9.1
>
> loaded via a namespace (and not attached):
> [1] ggplot2_0.8.9 tools_2.13.0
>
>
>
> http://lists.gnu.org/archive/html/igraph-help/2008-04/msg00049.html
>
>
> --
> --------------------------------------------------
> Michael M. Bishop, Ph.D. Candidate
> Department of Sociology
> University of Chicago
>
> _______________________________________________
> igraph-help mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/igraph-help
>



-- 
Gabor Csardi <address@hidden>     MTA KFKI RMKI