Seminaire: Alpage (Paola Merlo) lundi 30 novembre

[Thierry Hamon]

Date: Wed, 25 Nov 2009 17:13:08 +0100
From: bcrabbe at linguist.jussieu.fr
Message-ID: <1259165588.4b0d579466f6c at kmail.linguist.univ-paris-diderot.fr>
X-url: http://www.aclweb.org/anthology/P/P09/P09-1033.pdf
X-url: http://www.aclweb.org/anthology/W/W09/W09-1205.pdf
X-url: http://www.aclweb.org/anthology/W/W08/W08-2122.pdf


****************** Séminaire Alpage *******************

         Séminaire de l'école doctorale de Paris 7

Il s'agit du séminaire de recherche en linguistique informatique
organisé par l'équipe Alpage, Alpage est une équipe mixte Inria --
Paris 7 qui centre ses intérêts scientifiques autour de l'analyse
syntaxique automatique et du traitement du discours pour la langue
française.

Le séminaire se tiendra ce lundi de 14.00 à 16.30 en salle Rose à
l'antenne INRIA Italie
23 avenue d'Italie
75013 Paris

Toute personne intéressée est la bienvenue.

***********************************************************
Lundi 30 novembre 2009, Paola Merlo (University of Geneva, Department
of Linguistics, Computational Learning and Computational Linguistics
Group) nous parlera de :

Developing a semantic parser for French by partial transfer from
English

Résumé :

One of the currently debated topics in Natural Language Processing is
the problem of semantic role labelling. Given a sentence like ``I want
to reserve a train from Geneva to Paris'', how do we determine
automatically that ``from Geneva'' is the source of the trip and ``to
Paris'' is the destination? The solution to this problem lies at the
heart of all applications that require language understanding, such as
dialogue systems, question answering, or machine translation.

Recent successes of machine learning methods in statistical parsing
and lexical acquisition pave the way to a learning approach for this
problem too.  In this presentation, I will survey our recent work on
this topic. I will first briefly discuss which semantic role labelling
scheme is most appropriate for our approach to this task, comparing
PropBank to VerbNet along several dimensions. I will then illustrate a
probabilistic model of joint learning of syntactic dependency trees
and semantic role dependencies which achieves very competitive results
on current multi-lingual data sets.  Finally I will describe work in
progress on parsing French (trained on the treebank provided to us by
the ALPAGE team) and automatically transferring semantic role
annotation onto French dependencies trees to create semantic role
annotations for French.

Lectures suggérées :

P. Merlo and L. van der Plas (2009), "Abstraction and Generalisation
in Semantic Role Labels: PropBank, VerbNet or both?", Proceedings of
ACL-IJCNLP 2009,
Singapore.(http://www.aclweb.org/anthology/P/P09/P09-1033.pdf)

Gesmundo A., J. Henderson, P. Merlo and I. Titov (2009) "A Latent
Variable Model of Synchronous Syntactic-Semantic Parsing for Multiple
Languages" Proceedings of the Conference on Natural Language Learning
(CONLL 2009), Shared Task, Boulder,
CO.(http://www.aclweb.org/anthology/W/W09/W09-1205.pdf)

Henderson, J. Paola Merlo, Gabriele Musillo and Ivan Titov (2008) "A
Latent Variable Model of Synchronous Parsing for Syntactic and
Semantic Dependencies", CONLL 08 shared task
(http://www.aclweb.org/anthology/W/W08/W08-2122.pdf)

***********************************************************

-------------------------------------------------------------------------
Message diffuse par la liste Langage Naturel <LN at cines.fr>
Informations, abonnement : http://www.atala.org/article.php3?id_article=48
English version       : 
Archives                 : http://listserv.linguistlist.org/archives/ln.html
                                http://liste.cines.fr/info/ln

La liste LN est parrainee par l'ATALA (Association pour le Traitement
Automatique des Langues)
Information et adhesion  : http://www.atala.org/
-------------------------------------------------------------------------