BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Lexicographic text analysis using non-negative factorisation techn
 iques - Sinead Williamson
DTSTART:20060119T150000Z
DTEND:20060119T160000Z
UID:TALK4634@talks.cam.ac.uk
CONTACT:Phil Cowans
DESCRIPTION:Non-negative matrix factorisation (NNMF) is proposed as an alt
 ernative to Principle Component Analysis for use in latent semantic analys
 is of large text corpora. NNMF recognises the inherent non-negativity of l
 anguage data\, and has been shown to organise data into easily understood 
 concept groups of words. The representation of articles as a linear superp
 osition of such concept groups provides an intuitive method for comparison
  between articles. 
LOCATION:Ryle Seminar Room\, Cavendish Laboratory
END:VEVENT
END:VCALENDAR