BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Learning semantic roles from child-directed data: insights from a 
 probabilistic model of early language learning - Afra Alishahi\, Departmen
 t of Computational Linguistics and Phonetics\, Saarland University
DTSTART:20100427T150000Z
DTEND:20100427T163000Z
UID:TALK23543@talks.cam.ac.uk
CONTACT:Chris Cummins
DESCRIPTION:Semantic roles are a critical aspect of linguistic knowledge b
 ecause they indicate the relations of the participants in an event to the 
 main predicate. Experimental studies on children and adults show that both
  groups use associations between general semantic roles such as Agent and 
 Theme\, and grammatical positions such as Subject and Object\, even in the
  absence of familiar verbs. Other studies suggest that semantic roles evol
 ve over time\, and might best be viewed as a collection of verb-based or g
 eneral semantic properties. A usage-based account of language acquisition 
 suggests that general roles and their association with grammatical positio
 ns can be learned from the data children are exposed to\, through a proces
 s of generalization and categorization.  \n\nWe propose a probabilistic us
 age-based model of semantic role learning. Our model can acquire associati
 ons between the semantic properties of the arguments of an event\, and the
  syntactic positions that the arguments appear in. These probabilistic ass
 ociations enable the model to learn general conceptions of roles\, based o
 nly on exposure to individual verb usages\, and without requiring explicit
  labeling of the roles in the input. The acquired role properties are a go
 od intuitive match to the expected properties of various roles\, and are u
 seful in guiding comprehension in the model to the most likely interpretat
 ion in the face of ambiguity. The learned roles can also be used to select
  the correct meaning of a novel verb in an ambiguous situation. \n
LOCATION:GR-06/07\, English Faculty Building
END:VEVENT
END:VCALENDAR