*****  To join INSNA, visit http://www.sfu.ca/~insna/  *****

This is relatively easy to do
It is described in the paper by Steve and I that appeared in connections

See http://www.analytictech.com/borgatti/analyzing_clique_overlap.htm

You need to form the k-plex overlap matrix and then submit this to a
clustering routine to combine the k-plexes (details are in the paper)

Here is the way to do it (note that it works for any clique routine that
has overlaps and UCINET 6 now does this automatically for cliques)

1 Run k-plexes

2 Assuming you have left the output sets on their default then run
Data>Affiliations on KPlexset

The resultant matrix called affiliations will give have as the i,j th
entry the number of actors the ith k-plex has with the jth k-plex.

3 Submit this matrix to an hierarchical clustering routine to combine
the k-plexes.

4 You will need to refer back to the output from the k-plex routine to
determine the actors in the clustering you find.


In general I find this method to be by far the best way to deal with
overlap.



lawrence e raffalovich wrote:
>
> *****  To join INSNA, visit http://www.sfu.ca/~insna/  *****
>
> Hi all,
> I'm using Ucinet with interlocking directorates to locate k-plexes.
> Now, I'd like to combine k-plexes with substantial overlaps. This appears
> to be a straightforward type of task,  but I can't think of a way to
> do it (other than manually).  Suggestions would be greatly appreciated.
>
> TIA,
> --Larry
>
> Lawrence E. Raffalovich
> Associate Professor
>
> Department of Sociology               e-mail: [log in to unmask]
> University at Albany
> State University of New York          Voice: (518) 442-4456
> 1400 Washington Ave.                  Fax:   (518) 442-4936
> Albany, NY 12222
>
> _____________________________________________________________________
> SOCNET is a service of INSNA, the professional association for social
> network researchers (http://www.sfu.ca/~insna/). To unsubscribe, send
> an email message to [log in to unmask] containing the line
> UNSUBSCRIBE SOCNET in the body of the message.

--
Martin Everett
School of Computing and Mathematical Sciences
University of Greenwich
Old Royal Naval College
Park Row                                          Tel +44(0)20 8331 8716
Greenwich                                         FAX +44(0)20 8331 8665
London SE10 9LS
UK

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