Print

Print


*****  To join INSNA, visit http://www.insna.org  *****

<<<-------- Steve Eichert-------->>>

> I've been using the Newman Girvan algorithm [1] to identify communities
> within networks of individuals, however, I've been told by someone who I
> respect greatly that Newman Girvan isn't the best algorithm to use for
> identifying communities when dealing with human networks.  So, the
> question
> I have for the group is: what algorithm would you recommend for
> identifying
> communities when working with networks of people.

If you are not searching for a partition of the set of vertices you
can consider also the islands approach:
  http://www.springerlink.com/content/tp3w7237m4624462/fulltext.html
  http://vlado.fmf.uni-lj.si/pub/networks/doc/sunbelt/islands.pdf

For a fast (hierarchical) clustering algorithm see
  http://www.inma.ucl.ac.be/%7Eblondel/workshops/2008/files/batagelj.pdf

Both approaches are implemented in Pajek.

Vlado
-- 
Vladimir Batagelj, University of Ljubljana, FMF, Department of Mathematics
  Jadranska 19, 1000 Ljubljana, Slovenia
http://vlado.fmf.uni-lj.si

_____________________________________________________________________
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.insna.org). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.