[Prev][Next][Index]
[xtag-meeting]: XTAG meeting Monday Aug 12: More practice talks for COLING
-
To: xtag-meeting@cis.upenn.edu, clunch@cis.upenn.edu
-
Subject: [xtag-meeting]: XTAG meeting Monday Aug 12: More practice talks for COLING
-
From: CARLOS A PROLO <prolo@linc.cis.upenn.edu>
-
Date: Fri, 9 Aug 2002 18:02:39 -0400 (EDT)
-
Sender: owner-xtag@linc.cis.upenn.edu
On Monday, Aug 12, 10:30, at the XTAG meeting, IRCS fishbowl(room 478)
we will have David and Carlos giving practice talks for their COLING papers.
Titles/abstracts follow.
Carlos
===========================================================================
Recovering Latent Information in Treebanks
David Chiang and Daniel M. Bikel
Many recent statistical parsers rely on a preprocessing step
which uses hand-written, corpus-specific rules to augment the training data
with extra information. For example, head-finding rules are used to augment
node labels with lexical heads. In this paper, we provide machinery to reduce
the amount of human effort needed to adapt existing models to new corpora:
first, we propose a flexible notation for specifying these rules that would
allow them to be shared by different models; second, we report on an experiment
to see whether we can use Expectation-Maximization to automatically fine-tune a
set of hand-written rules to a particular corpus.
===========================================================================
Generating the XTAG English grammar using metarules
Carlos A. Prolo
We discuss a grammar development process used to generate the trees
of the wide-coverage Lexicalized Tree Adjoining Grammar (LTAG) for English
of the XTAG Project. Result of the coupling of Becker's metarules
and a simple yet principled hierarchy of rule application,
the approach has been successful to generate the large set of verb trees
in the grammar, from a very small initial set of trees.