[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [igraph] cohesive.blocks()

From: Gabor Csardi
Subject: Re: [igraph] cohesive.blocks()
Date: Mon, 28 Apr 2008 16:30:35 +0200
User-agent: Mutt/1.5.13 (2006-08-11)


to be honest, i don't really understand the question. 
In your graph (==block 1, cohesion 1) the algorithm finds 
two sub-blocks (block 2, cohesion 2, and block 3, cohesion 2).
In block 2 there is a (more) cohesive sub-block, this 
will be block 4, it has cohesion 5. Within block 3, there is 
a 3-cohesive block, this will be block 5. 

In the last case, the remainder of block 3, i.e. the vertices 
not in block 5 (vertices 8, 9, 10, 11, 12, 14, 15) do not 
constitute a cohesive block, that's why they are not included in 
the result.

If in a B cohesive block there is a sub-block C with higher 
cohesion value, that does not mean that B-C (the remainder)
has a higher cohesion value (than the original B block) as well.

Does this answer your question?

On Mon, Apr 28, 2008 at 05:54:20PM +0900, MATSUDA, Noriyuki wrote:
> Hello:
>    I have some difficulty in relating the performance of   
> cohesive.blocks() to the original paper by Moody and White (2003),  
> particularly the concepts of components and blocks.  Here is my trial  
> on Figures 2 and 3 in the paper:
>   It separates terminal (bottom) blocks 4 and 5, leaving  {17-23}  
> unextracted.   Do we need to identify {17-23} by taking the  
> difference between blocks 2 and 4?  If so, it would require a lot of  
> care even with a graph of medium size.  Or have I made errors in  
> specifying edges?
>   Thanks in advance for helping me out.

Csardi Gabor <address@hidden>    UNIL DGM

reply via email to

[Prev in Thread] Current Thread [Next in Thread]