| View previous topic :: View next topic |
| Author |
Message |
Sengly
Guest
|
 Posted: Tue Jul 15, 2008 8:40 am Post subject: Wordnet's most frequent sense of a term |
 |
|
Dear all,
I am just starting using Wordnet (via NLTK API) and I would like to
know how to get the most frequent sense of a term?
Please share your experience.
Thank you.
Best regards,
Sengly |
|
| |
|
 Back to top |
Keenlearner
Guest
|
| Posted: Thu Jul 17, 2008 1:53 pm Post subject: Re: Wordnet's most frequent sense of a term |
|
|
| Quote: | I am just starting using Wordnet (via NLTK API) and I would like to
know how to get the most frequent sense of a term?
|
I have just a few weeks of experience with NLTK then I moved to Perl,
| Quote: | Please share your experience.
as far as I know of, Perl is the most friendly language to me in NLP |
research, consider its close relation with text processing. Moreover
there are a lot of NLP module out there in CPAN
I don't know much about NLTK API, but you can do this in Perl module
WordNet::QueryData at http://search.cpan.org/~jrennie/WordNet-QueryData-1.47/QueryData.pm
How you define "term" ? Do you mean to get the most frequent sense of
a word such as love#n ? If so you can query with that module using
Perl as
use WordNet::QueryData;
$wn = new WordNet::QueryData;
@syns = $wn->querySense("love#n", "syns");
print $syns[0];
Regards,
William Kisman |
|
| |
|
| Back to top |
Brian Martin
Guest
|
| Posted: Thu Jul 17, 2008 8:02 pm Post subject: Re: Wordnet's most frequent sense of a term |
|
|
Hi,
I'm not an expert, but I think the most frequent sense of a term in
Wordnet is listed first. I think they are listed in descending freq order.
So whatever language you use, the most frequent sense is the first one
i.e. $syns[0] in the example below.
Personally, I'm using Python, but not using NLTK, just direct reading of
the Wordnet files and other data sources.
regards,
Brian
Keenlearner wrote:
| Quote: | I am just starting using Wordnet (via NLTK API) and I would like to
know how to get the most frequent sense of a term?
I have just a few weeks of experience with NLTK then I moved to Perl,
Please share your experience.
as far as I know of, Perl is the most friendly language to me in NLP
research, consider its close relation with text processing. Moreover
there are a lot of NLP module out there in CPAN
I don't know much about NLTK API, but you can do this in Perl module
WordNet::QueryData at http://search.cpan.org/~jrennie/WordNet-QueryData-1.47/QueryData.pm
How you define "term" ? Do you mean to get the most frequent sense of
a word such as love#n ? If so you can query with that module using
Perl as
use WordNet::QueryData;
$wn = new WordNet::QueryData;
@syns = $wn->querySense("love#n", "syns");
print $syns[0];
Regards,
William Kisman
|
|
|
| |
|
| Back to top |
|
