BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Experiments with Non-parametric Topic Models - Prof. Wray Buntine 
 (Monash University)
DTSTART:20150123T110000Z
DTEND:20150123T120000Z
UID:TALK56842@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:This talk will cover some of our recent work in extended topic
  models\nto serve as tools in text mining and NLP (and hopefully\, later\,
  in IR)\nwhen some semantic analysis is required.  In some sense our goals
  are\nakin to the use of Latent Semantic Analysis.  The basic\ntheoretical
 /algorithmic tool we have for this is non-parametric\nBayesian methods for
  reasoning on hierarchies of probability vectors.\nThe concepts will be in
 troduced but not the statistical detail. Then\nI'll present some of our KD
 D 2014 paper (Experiments with\nNon-parametric Topic Models) that is curre
 ntly the best performing topic\nmodel by a number of metrics\, and some ex
 tended work such as\n"Bibliographic Analysis with the Citation Network Top
 ic Model" (ACML 2014)\nand "Topic Segmentation with a Structured Topic Mod
 el"\n(NAACL 2013).  Various evaluations and comparisons will be made.\n\nB
 io:\n\nProf. Wray Buntine joined Monash University in February 2014 after 
 7\nyears at NICTA in Canberra Australia.  He was previously of Helsinki\nI
 nstitute for Information Technology from 2002\, and at NASA Ames\nResearch
  Center\, University of California\, Berkeley\, and Google. He is\nknown f
 or his theoretical and applied work in document and text\nanalysis\, data 
 mining and machine learning\, and probabilistic methods.\nHe applies proba
 bilistic and non-parametric methods to tasks such as\ntext analysis.  In 2
 009 he was programme co-chair of ECML-PKDD in\nBled\, Slovenia\, and was p
 rogramme co-chair of ACML in Singapore in\n2012.  He reviews for conferenc
 es such as ACML\, ECIR\, SIGIR\,\nECML-PKDD\, ICML\, NIPS\, UAI\, and KDD\
 , and is on the editorial board of\nData Mining and Knowledge Discovery\n
LOCATION:Engineering Department\, CBL Room BE-438.
END:VEVENT
END:VCALENDAR
