BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Appellation d'Origine Contrôlée - Language variability: a major 
 challenge for natural language applications - Thierry Poibeau\, LIPN-CNRS\
 , Paris\; RCEAL affiliated lecturer
DTSTART:20090505T150000Z
DTEND:20090505T163000Z
UID:TALK17817@talks.cam.ac.uk
CONTACT:Susan Rolfe
DESCRIPTION:Many Natural Language Processing (NLP) applications are challe
 nged by the great variability of language – the fact that the same word 
 may refer to various things and that the same idea could be expressed in a
  variety of ways. This talk will focus on language variability and\, in pa
 rticular\, on how existing NLP techniques handle it. \n\n\nI will first fo
 cus on named entities\, i.e. sequences of text corresponding to person nam
 es\, location names\, dates\, currencies\, etc. Named entities are importa
 nt for text processing since they are good indicators of the content of te
 xts and can serve as a basis for deeper analysis. They are typically consi
 dered as "rigid designators"\, unambiguously referring to a single\, stabl
 e entity in the world. I will show that this assumption is not always corr
 ect\; rather\, the meaning of a named entity can be affected by context. I
  will illustrate this with the case of metonymy and show that although met
 onymy is a relatively well-understood linguistic phenomenon\, it is diffic
 ult to analyse it using a fully automatic approach. \n\n\nI will then focu
 s on discourse processing\, on a task which aims to automatically structur
 e free text according to a set of semantic principles. Automatic discourse
  analysis is challenging since it requires considering multiple linguistic
  cues and their interaction in complex patterns. These patterns may includ
 e conflicting information among which the parser has to choose. I will pre
 sent a framework particularly designed to choose an optimal solution from 
 a range of complex\, interacting constraints that sometimes contradict the
 mselves. This approach is implemented and evaluated for Health Practices G
 uidelines (i.e. short documents describing the practices that physicians s
 hould follow).  \n\n\nIn the conclusion\, I will discuss the dependency of
  computational approaches on language usage: the meaning of a linguistic i
 tem is largely dependent on context\, and the context is difficult model i
 n advance.\n
LOCATION:GR-06/07\, English Faculty Building
END:VEVENT
END:VCALENDAR