BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Analysing biomedical text with the Stanford dependency grammar - A
 ndrew Clegg\, Birkbeck\, University of London
DTSTART:20071019T110000Z
DTEND:20071019T120000Z
UID:TALK8703@talks.cam.ac.uk
CONTACT:Johanna Geiss
DESCRIPTION:The Stanford parser and natural language tools come with a dep
 endency grammar and a deterministic algorithm for mapping from standard Pe
 nn Treebank-style parse trees to dependency graphs. We have shown how this
  provides a lingua franca for comparing various statistical parsers on bio
 medical text\, using benchmarking methods that highlight deficiencies in t
 heir behaviour much better than tree-based analyses. We also believe depen
 dency graphs offer a tractable intermediate step between syntax and semant
 ics and have developed some simple approaches to information extraction ba
 sed on graph traversal.
LOCATION:SW01 Computer Laboratory
END:VEVENT
END:VCALENDAR