Re: [igraph] interpretation of hub/auth scores

[Tamás Nepusz]

Hi Matthew,  

Yes, that’s odd indeed. Let me double-check our code; I suspect an ARPACK 
convergence problem here but I’m not sure (and I don’t know why there’s no 
warning if ARPACK indeed fails to converge).  

--  
T.


On Monday, 2 December 2013 at 21:30, Matthew Galati wrote:

> I am considering this small directed graph.  
>  
> The results score node 1 as the dominate hub and the rest of the nodes with 
> value 0 (i.e., no' hubness'). This seems odd to me. For example, node 2, has 
> outlinks to 3 different nodes (1,4, and 7). So, I would expect it to have 
> some relative level of hubness.  
>  
> > g <- graph.empty()  
> > g <- g+vertices(1,2,3,4,5,6,7)
> > g <- g+edge("1","2")
> > g <- g+edge("1","3")
> > g <- g+edge("1","5")
> > g <- g+edge("1","6")
> > g <- g+edge("2","1")
> > g <- g+edge("2","4")
> > g <- g+edge("3","1")
> > g <- g+edge("5","1")
> > g <- g+edge("2","7")
> > E(g)
>  
> Edge sequence:
>  
> [1] 1 -> 2
> [2] 1 -> 3
> [3] 1 -> 5
> [4] 1 -> 6
> [5] 2 -> 1
> [6] 2 -> 4
> [7] 3 -> 1
> [8] 5 -> 1
> [9] 2 -> 7
>  
> > hub.score(g,scale=FALSE)$vector
>  
> 1 2 3 4 5 6 7  
> 0.5733822 0.0000000 0.0000000 0.0000000 0.0000000 0.0000000 0.0000000  
>  
>  
> _______________________________________________
> igraph-help mailing list
> address@hidden (mailto:address@hidden)
> https://lists.nongnu.org/mailman/listinfo/igraph-help