BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A Simple Unsupervised Learner for POS Disambiguation Rules Given O
 nly a Minimal Lexicon. - Yue Zhang
DTSTART:20101122T124500Z
DTEND:20101122T134500Z
UID:TALK28071@talks.cam.ac.uk
CONTACT:Jimme Jardine
DESCRIPTION:Yue will present the following paper:\n\nA Simple Unsupervised
  Learner for POS Disambiguation Rules Given Only a Minimal Lexicon.\nQiuye
  Zhao and Mitch Marcus. ACL 2009\nhttp://www.aclweb.org/anthology/D/D09/D0
 9-1072.pdf\n\nWe propose a new model for unsupervised POS tagging based on
  linguistic distinctions between open and closed-class items. \nExploiting
  notions from current linguistic theory\, the system uses far less informa
 tion than previous systems\, far simpler computational methods\, and far s
 parser descriptions in learning contexts. By applying simple language acqu
 isition techniques based on counting\, the system is given the closed-clas
 s lexicon\, acquires a large open-class lexicon and then acquires disambig
 uation rules for both. This system achieves a 20% error reduction for POS 
 tagging over state-of-the-art unsupervised systems tested under the same c
 onditions\, and achieves comparable accuracy when trained with much less p
 rior information.\n\n\nCheers\,\nJimme\n
LOCATION:GS15\, Computer Laboratory
END:VEVENT
END:VCALENDAR