BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Aggregating Minds - Dr Yoram Bachrach\, Microsoft Research Cambrid
 ge
DTSTART:20121030T160000Z
DTEND:20121030T170000Z
UID:TALK40911@talks.cam.ac.uk
CONTACT:Luning Sun
DESCRIPTION:How can we combine crowds of people into one giant mind in ord
 er to solve a problem at hand? Given many individual opinions regarding a 
 question\, how can we identify both the correct answer and the strongest m
 embers of the crowd? Crowdsourced solutions to various problems have prove
 n to be of extremely high quality in many business situations. However\, i
 n many such situations\, finding the best solution is equivalent to gradin
 g the tests of several students\, without knowing the correct answers to t
 he questions. I will show how machine learning techniques based on item-re
 sponse theory can overcome this problem\, and discuss some applications of
  this model.
LOCATION:Seminar Room\, Department of Psychology\, Downing Site\, Cambridg
 e
END:VEVENT
END:VCALENDAR