BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Predicting Strong Associations on the Basis of Corpus Data - Colin
  Kelly (Computer Laboratory)
DTSTART:20090302T123000Z
DTEND:20090302T133000Z
UID:TALK17185@talks.cam.ac.uk
CONTACT:Diarmuid Ó Séaghdha
DESCRIPTION:At this session of the NLIP Reading Group we'll be discussing 
 the following paper:\n\nYves Peirsman and Dirk Geeraerts. 2009. Predicting
  Strong Associations on the Basis of Corpus Data. In Proceedings of the 12
 th Conference of the European Chapter of the Association for Computational
  Linguistics (EACL-09). \n\nThis paper doesn't seem to be up on the Web ye
 t\; please contact the group organiser if you need a copy.\n\n*Abstract:*\
 nCurrent approaches to the prediction of associations rely on just one typ
 e of information\, generally taking the form of either word space models o
 r collocation\nmeasures. At the moment\, it is an open question how these 
 approaches compare to one another. In this paper\, we will investigate the
  performance of these two\ntypes of models and that of a new approach base
 d on compounding. The best single predictor is the log-likelihood ratio\, 
 followed closely by the document-based word space model. We will show\, ho
 wever\,\nthat an ensemble method that combines these two best approaches w
 ith the compounding algorithm achieves an increase in performance of almos
 t 30% over the current state of the art.
LOCATION:GS15\, Computer Laboratory
END:VEVENT
END:VCALENDAR