BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:BioCaster 2.0: Online text analysis for early alerting of disease 
 outbreaks - Nigel Collier - National Institute of Informatics\, Tokyo
DTSTART:20101027T110000Z
DTEND:20101027T120000Z
UID:TALK27552@talks.cam.ac.uk
CONTACT:Thomas Lippincott
DESCRIPTION:Accurate and timely detection of infectious disease outbreak e
 vents of international concern is necessary to help support risk assessmen
 t and ultimately to save lives and livelihoods. In this talk I will introd
 uce high throughput biomedical fact extraction from the 'grey' literature 
 - the large body of non- peer reviewed texts on the Web and consider how w
 e might meet the challenge of deciding which events are 'unusual'. Modelin
 g 'norms' and detecting their violations raises several key issues which w
 ill be touched on in the talk: (1) how to bridge the ambiguity gap between
  laymen's terms and extant biomedical ontological resources\, (2) how to i
 nterpret evidence across documents\, (3) how to achieve real time processi
 ng of thousands of documents per hour\, (4) how we might provide a realist
 ic benchmark standard for alerting. 
LOCATION:SW01\, Computer Laboratory
END:VEVENT
END:VCALENDAR