\magnification=1200
\baselineskip=20pt
\nopagenumbers
\font\big=cmr12 scaled \magstep2

\centerline{\bf STANFORD UNIVERSITY}
\centerline{\bf DEPARTMENT OF STATISTICS}
\centerline{\big DEPARTMENTAL SEMINAR}
\bigskip

\baselineskip=12pt
\centerline{4:15 p.m., Tuesday, May 1, 2001}
\centerline{Sequoia Hall Rm. 200}
\centerline{(Cookies at 3:45 in 1st Floor Lounge)}

\bigskip
\baselineskip=15pt
\centerline{\sl Christopher Manning}
\centerline{\sl Computer Science and Linguistics}
\centerline{\sl Stanford University}
\bigskip
\centerline{\bf Probabilistic Head-driven Parsing}
\bigskip

A central problem in natural language processing is resolving sentence
structure ambiguities, and this has been approached in recent years by
learning probabilistic models from hand-parsed sentences.  This paper
discusses a particular approach of this sort, which builds on two
central linguistic notions: a phrasal head is a locus of constraining
information, and that around the head is an "island of certainty" --
verb arguments, adjectival modifiers, and the like.  A parsing model
that works outwards from heads has several advantages: the model is
immune to some fairly arbitrary aspects of tree geometry, it can easily
capture important length and distance-based effects, and the model can
more cleanly capture the intuitions of dependency-based approaches to
grammar.

\bye