BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A Hierarchical Bayesian Language Model based on Pitman-Yor Process
 es\n - Hanna Wallach\, University of Cambridge
DTSTART:20060824T090000Z
DTEND:20060824T100000Z
UID:TALK5210@talks.cam.ac.uk
CONTACT:Hanna Wallach
DESCRIPTION:"Paper":http://www.cs.berkeley.edu/~ywteh/research/bayesnlp/ac
 l2006.pdf\n(also "Tech. report":http://www.cs.berkeley.edu/~ywteh/research
 /bayesnlp/hpylm.pdf)\n\nWe propose a new hierarchical Bayesian n-gram mode
 l of natural languages. Our model makes use of a generalization of the com
 monly used Dirichlet distributions called Pitman-Yor processes which produ
 ce power-law distributions more closely resembling those in natural langua
 ges. We show that an approximation to the hierarchical Pitman-Yor language
  model recovers the exact formulation of interpolated Kneser-Ney\, one of 
 the best smoothing methods for n-gram language models.  Experiments verify
  that our model gives cross entropy results superior to interpolated Knese
 r-Ney and comparable to modified Kneser-Ney.\n
LOCATION:Room 911\, Rutherford Building\, Cavendish Laboratory\, Departmen
 t of Physics
END:VEVENT
END:VCALENDAR