I would like to make a small correction in the
following:
> The partitional methods included with cluto are
> repeated bisections (rb),
> repeated bisections with global optimization (rbr),
> and direct (direct)
> clustering. A basic rule of thumb is that rb or rbr
> are good choices,
> unless you have a very small amount of data, in
> which case direct is
> good. By small, I mean maybe 20 contexts to cluster.
Use of "direct" clustering method is recommended for
datasets with k (#clusters) less than 10-20 and not
for smaller datasets.
As I understand, out of the 3 partitional methods
supported by Cluto, namely - rb, rbr and direct - the
direct clustering method is the only purely
partitional method while the other 2 (rb & rbr) are
kind of hybrid methods (partitional + hierarchical).
The direct method splits the given dataset into k
clusters in just one step while the rb and rbr methods
reach the desired number of clusters by repeatedly
splitting one of the clusters into two.
Thanks,
Anagha
__________________________________
Yahoo! Mail - PC Magazine Editors' Choice 2005
http://mail.yahoo.com
-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. Download
it for free - -and be entered to win a 42" plasma tv or your very own
Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php
_______________________________________________
senseclusters-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/senseclusters-users
