BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Making Computer Science more Social: Speed Dating and the History 
 of Science - Dan Jurafsky\, Stanford University
DTSTART:20091126T170000Z
DTEND:20091126T180000Z
UID:TALK21629@talks.cam.ac.uk
CONTACT:Laura Rimell
DESCRIPTION:This talk describes two ongoing projects in our attempt to app
 ly natural language processing to social science tasks like history and so
 cial psychology. We first describe a system for detecting human social int
 entions from spoken conversation: whether a speaker is awkward\, friendly\
 , or flirtatious. We create and use a new spoken corpus of 991 4-minute sp
 eed-dates. Using rich dialogue\, lexical\, disfluency\, and prosodic featu
 res\, we are able to detect flirtatious\, awkward\, and friendly styles in
  noisy natural conversational data with above 70% accuracy\, significantly
  outperforming the human interlocutors. We suggest that humans are very po
 or perceivers of flirtatiousness or friendliness in others\, instead often
  projecting their own intended behavior onto their interlocutors. (Joint w
 ork with Dan McFarland\, Education\, and Rajesh Ranganath\, Computer Scien
 ce). We then describe our results on detecting how ideas spread in the his
 tory of science\, by studying what factors lead scholars to cite one anoth
 er and by tracing the spread of scientific topics from subfield to subfiel
 d in research paper anthologies. (Joint work with Steven Bethard\, David H
 all\, and Chris Manning). 
LOCATION:Lecture Theatre 1\, Computer Laboratory
END:VEVENT
END:VCALENDAR