You are viewing a plain text version of this content. The canonical link for it is here.
Posted to java-user@lucene.apache.org by Jens Grivolla <j-...@grivolla.net> on 2007/07/25 21:04:31 UTC
MoreLikeThis for multiple documents
Hello,
I'm looking to extract significant terms characterizing a set of
documents (which in turn relate to a topic).
This basically comes down to functionality similar to determining the
terms with the greatest offer weight (as used for blind relevance
feedback), or maximizing tf.idf (as is done in MoreLikeThis).
Is there anything like this already implemented, or do I need to iterate
through all documents in the set "manually", re-tokenize each one (or
maybe use TermVectors), and then calculate the weight for each term?
Thanks,
Jens
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org
Re: MoreLikeThis for multiple documents
Posted by Grant Ingersoll <gs...@apache.org>.
I have some sample code for doing relevance feedback across multiple
documents at http://www.cnlp.org/apachecon2005
It could be modified to provide more of the MoreLikeThis
functionality (i.e. determining important terms via tf/idf) for now
it just takes the top X terms
-Grant
On Jul 25, 2007, at 3:04 PM, Jens Grivolla wrote:
> Hello,
>
> I'm looking to extract significant terms characterizing a set of
> documents (which in turn relate to a topic).
>
> This basically comes down to functionality similar to determining
> the terms with the greatest offer weight (as used for blind
> relevance feedback), or maximizing tf.idf (as is done in
> MoreLikeThis).
>
> Is there anything like this already implemented, or do I need to
> iterate through all documents in the set "manually", re-tokenize
> each one (or maybe use TermVectors), and then calculate the weight
> for each term?
>
> Thanks,
> Jens
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
> For additional commands, e-mail: java-user-help@lucene.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org
Re: MoreLikeThis for multiple documents
Posted by Mathieu Lecarme <ma...@garambrogne.net>.
Jens Grivolla a écrit :
> Hello,
>
> I'm looking to extract significant terms characterizing a set of
> documents (which in turn relate to a topic).
>
> This basically comes down to functionality similar to determining the
> terms with the greatest offer weight (as used for blind relevance
> feedback), or maximizing tf.idf (as is done in MoreLikeThis).
>
> Is there anything like this already implemented, or do I need to
> iterate through all documents in the set "manually", re-tokenize each
> one (or maybe use TermVectors), and then calculate the weight for each
> term?
http://project.carrot2.org/index.html may be your friend.
M.
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org