[CL+NLP Lunch] Joint ML+NLP Lunch, Miguel Ballesteros, Monday October 6th @ 12:00pm

[Dallas Card]

Please join us for a special joint ML+NLP lunch at noon on October 6th, where
Miguel Ballesteros will be speaking about dependency parsing. Lunch will
be provided!

ML+NLP lunch <http://www.cs.cmu.edu/~nlp-lunch/>
Monday, October 6th at 12:00pm
GHC 6115


Speaker: Miguel Ballesteros, Visiting Lecturer / Postdoc at Universitat
Pompeu Fabra

*Title: Going to the Roots of Dependency Parsing*

In this seminar I will first introduce transition-based dependency parsing
and present the
conclusions extracted from a journal paper that I have never had the
chance to present in public,
besides I'm going to sum up my current, past and future research
collaboration projects with some
new results and developments.
--
Dependency trees used in syntactic parsing often include a root node
representing a dummy
word prefixed or suffixed to the sentence, a device that is generally
considered a mere technical
convenience and is tacitly assumed to have no impact on empirical results.
We demonstrate that
this assumption is false and that the accuracy of data-driven dependency
parsers can in fact be
sensitive to the existence and placement of the dummy root node. In
particular, we show that
a greedy, left-to-right, arc-eager transition-based parser consistently
performs worse when the
dummy root node is placed at the beginning of the sentence (following the
current convention
in data-driven dependency parsing) than when it is placed at the end or
omitted completely.
Control experiments with an arc-standard transition-based parser and an
arc-factored graph-
based parser reveal no consistent preferences but nevertheless exhibit
considerable variation in
results depending on root placement. We conclude that the treatment of
dummy root nodes in
data-driven dependency parsing is an underestimated source of variation in
experiments and
may also be a parameter worth tuning for some parsers.


Miguel is a Visiting lecturer - Postdoc in Pompeu Fabra University,
Barcelona, Spain. He works on natural language processing and machine
learning with a special interest on linguistic structure prediction
problems, such as dependency parsing and phrase structure parsing.  He
completed his BsC,  MsC and PhD at the Universidad Complutense de Madrid.
During the last years, he was a Visiting Researcher in Universities of
Uppsala,  Birmingham and Singapore.


-- 
Dallas Card
Machine Learning Department
Carnegie Mellon University