Evaluating Word Sense Disambiguation Systems
There are now many computer programs for automatically determining which
sense a word is being used in. One would like to be able to say which were
better, which worse, and also which words, or varieties of language, presented
particular problems to which programs. SENSEVAL is designed to meet this
need.
The first SENSEVAL took place in the summer of 1998, for English, French
and Italian, culminating in a workshop held at Herstmonceux Castle, Sussex,
England on September 2-4. The workshop for the SENSEVAL-2 will be held
in conjunction with ACL/EACL 2001, in Toulouse, France, on July
6 and 7 2001.
SENSEVAL 2
This will take place in the first half of 2001. The timetable is available
as ASCII text here or in postscript
here.
For people interested in setting up exercises for additional languages,
the key issues are sketched here
(postscript).
Pointers
-
Schedule, SENSEVAL-2 ascii or ps
-
Task design for SENSEVAL-2
-
SENSEVAL-2:
senseval-discuss mailing list archive
-
Home page for SENSEVAL-1.
-
English SENSEVAL resources now in the public
domain.
-
For the 1998 French and Italian tasks see the ROMANSEVAL
site.
-
A full account of SENSEVAL will be appearing in a Special Issue of the
Journal Computers and the Humanities (34(1-2)), guest editors Kilgarriff
and Palmer, in Spring 2000.
-
SENSEVAL: An Exercise in Evaluating Word Sense
Disambiguation Programs A. Kilgarriff. In Proc. LREC, Granada,
May 1998. Pp 581--588. Postscript, 6 pages.
-
English SENSEVAL: Report and Results A.
Kilgarriff and J. Rosenzweig. To appear in Proc. LREC, Athens, May-June
2000.
-
Summaries of the EPSRC-funded component of the work can be found here:
1
page, in html or 5 pages, in postscript.
This page is maintained by the SENSEVAL
coordinator and was last updated on 16 May 2000 .
