Re: Corpora: Discovering new senses

Ken Litkowski (ken@clres.com)
Tue, 23 Nov 1999 12:02:50 -0500

"Dimitris K." wrote:
>
> Sense tagging might have been a bottleneck for such work in the past
> but since there is much work going on, around sense disambiguation,
> are there any recent attempts to use sense tagging (or other
> methodologies)
> for the discovery of new senses?
>

Sorry, but this is still a bottleneck.

> I am interested in work related to discovering/identifying NEW
> senses in corpora, possibly with the help of existing monolingual MRDs
> such as LDOCE (or even large ontologies).

We're not there yet. I think we still haven't sussed out the structure
of MRDs sufficiently. MindNet possibly is the furthest along in this
respect.

> Note that I am *not* interested on classifying NEW words in the sense
> of, for instance Dekang Lin's work in :
>

But, Dekang's work is crucial to identifying new senses, primarily by
comparing his results against existing senses and then by examining the
subcat patterns and lexical preferences for ones that don't fit what is
indicated in current MRDs. His work is necessary infrastructure. We
don't yet have the wherewithal to exploit his work.

Us computational lexicologists are still chugging along.

Ken

-- 
Ken Litkowski                     TEL.: 301-482-0237
CL Research                       EMAIL: ken@clres.com
9208 Gue Road
Damascus, MD 20872-1025 USA       Home Page: http://www.clres.com