Thank you so much for your explanation.
On 2 June 2015 at 17:31, Alessandro Benedetti
wrote:
> The scope in there is to try to make clustering lighter and more related to
> the query.
> The summary produced is a fragment that is surrounding the query terms in
> the document content.
> Actually t
The scope in there is to try to make clustering lighter and more related to
the query.
The summary produced is a fragment that is surrounding the query terms in
the document content.
Actually this is arguably a way to improve the quality of clusters, but for
sure it makes the clustering operation l
Thank you so much Alessandro.
But i do not find any difference with the quality of the clustering results
when I change the hl.fragszie to a even though I've set my
carrot.produceSummary to true.
Regards,
Edwin
On 1 June 2015 at 17:31, Alessandro Benedetti
wrote:
> Only to clarify the initi
Only to clarify the initial mail, The carrot.fragSize has nothing to do
with the number of clusters produced.
When you select to work with field summary ( you will work only on snippets
from the original content, snippets produced by the highlight of the query
in the content), the fragSize will sp
Thank you Stanislaw for the links. Will read them up to better understand
how the algorithm works.
Regards,
Edwin
On 29 May 2015 at 17:22, Stanislaw Osinski <
stanislaw.osin...@carrotsearch.com> wrote:
> Hi,
>
> The number of clusters primarily depends on the parameters of the specific
> cluster
Hi,
The number of clusters primarily depends on the parameters of the specific
clustering algorithm. If you're using the default Lingo algorithm, the
number of clusters is governed by
the LingoClusteringAlgorithm.desiredClusterCountBase parameter. Take a look
at the documentation (
https://cwiki.a
