BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Context-Enhanced Citation Sentiment Detection - Awais Athar\, Univ
 ersity of Cambridge
DTSTART:20120525T110000Z
DTEND:20120525T113000Z
UID:TALK38342@talks.cam.ac.uk
CONTACT:Ekaterina Kochmar
DESCRIPTION:Sentiment analysis of citations in scientific papers and artic
 les is a new\nand interesting problem which can open up many exciting new 
 applications in\nbibliographic search and bibliometrics. Current work on c
 itation sentiment\ndetection focuses on only the citation sentence. In thi
 s paper\, we address\nthe problem of context-enhanced citation sentiment d
 etection. We present a\nnew citation sentiment corpus which has been annot
 ated to take the dominant\nsentiment in the entire citation context into a
 ccount. We believe that this\ngold standard is closer to the truth than an
 notation that looks only at the\ncitation sentence itself. We then explore
  the effect of context windows of\ndifferent lengths on the performance of
  a state-of-the-art citation\nsentiment detection system when using this c
 ontext-enhanced gold standard\ndefinition.
LOCATION:FW26\, Computer Laboratory
END:VEVENT
END:VCALENDAR