BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:NLIP reading group: Global learning of typed entailment rules - Ma
 rek Rei (University of Cambridge)
DTSTART:20111117T121500Z
DTEND:20111117T130000Z
UID:TALK34474@talks.cam.ac.uk
CONTACT:Jimme Jardine
DESCRIPTION:Marek will be talking about the following:\n\nGlobal learning 
 of typed entailment rules\nJonathan Berant\, Ido Dagan\, Jacob Goldberger\
 nWinner of best student paper at ACL 2011.\nPDF: http://www.aclweb.org/ant
 hology/P/P11/P11-1062.pdf\n\nAbstract: Extensive knowledge bases of entail
 ment rules between predicates are crucial for applied semantic inference. 
 In this paper we propose an algorithm that utilizes transitivity constrain
 ts to learn a globally-optimal set of entailment rules for typed predicate
 s. We model the task as a graph learning problem and suggest methods that 
 scale the algorithm to larger graphs. We apply the algorithm over a large 
 data set of extracted predicate instances\, from which a resource of typed
  entailment rules has been recently released (Schoenmackers et al.\, 2010)
 . Our results show that using global transitivity information substantiall
 y improves performance over this resource and several baselines\, and that
  our scaling methods allow us to increase the scope of global learning of 
 entailment-rule graphs.\n
LOCATION:GS15\, Computer Laboratory
END:VEVENT
END:VCALENDAR