<?xml version="1.0" standalone="yes"?>
<Paper uid="C00-1051">
  <Title>Committee-based Decision Making in Probabilistic Partial Parsing</Title>
  <Section position="3" start_page="0" end_page="348" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> There have been a number of attempts to use statistical techniques to improve parsing performance. While this goal has been achieved to a certain degree given the increasing availability of large tree banks, the remaining room tbr the improvement appears to be getting saturated as long as only statistical techniques are taken into account. This paper explores two directions tbr the next step beyond the state of the art of statistical parsing: probabilistic partial parsing and committee-based decision making.</Paragraph>
    <Paragraph position="1"> Probabilistic partial parsing is a probabilistic extension of the existing notion of partial parsing ( e.g. (Jensen et al., 1993)) where a parser selects as its output only a part of the parse tree that are probabilistically highly reliable. This decision-making scheme enables a fine-grained arbitrary choice on the trade-off between accuracy and coverage. Such trade-oil is important since there are various applications that require reasonably high accuracy even sacrificing coverage. A typical example is the t)araI)hrasing task embedded in summarization, sentence simplification (e.g. (Carroll et al., 1998)), etc.</Paragraph>
    <Paragraph position="2"> Enabling such trade-off&amp;quot; choice will make state-of the-art parsers of wider application. Partial parsing has also been proven useflll ibr bootstrapping leanfing.</Paragraph>
    <Paragraph position="3"> One may suspect that the realization of partial parsing is a trivial matter in probabilistic parsing just because a probabilistic parser inherently has the notion of &amp;quot;reliability&amp;quot; and thus has the trade-off:' between accuracy and coverage. However, there has so far been surprisingly little research focusing on this matter and ahnost no work that evaluates statistical parsers according to their coverage-accuracy (or recallprecision) curves. Taking the significance of partial parsing into account, therefi)re in this paper, we evaluate parsing perfbrmance according tO coverage-accuracy cnrves.</Paragraph>
    <Paragraph position="4"> Committee-based decision making is to con&gt; bine the outputs from several difl'erent systems (e.g. parsers) to make a better decision. Recently, there have been various attempts to at)ply committee-based techniques to NLP tasks such as POS tagging (Halteren et al., 1998; Brill et al., 1998), parsing (Henderson and Brill, 1999), word sense disambiguation (Pedersen, 2000), machine translation (lh'ederking and Nirenburg, 1994), and speech recognition (Fiscus, 1997). Those works empirically demonstrated that combining different systems often achieved significant improvelnents over the previous best system.</Paragraph>
    <Paragraph position="5"> In order to couple those committee-based  schemes with t)robat)ilistic t)artial parsing, however, Olle would still need to make a fllrther extension. Ainling at this coupling, ill this t)at)er, we consider a general framework of (:ommil, tee-based decision making that consists of ~ set of weighting flmctions mid a combination flmction, and (lis('uss how that Kalnework enal)les the coupling with t)robal)ilistic t)artial t)m:sing. To denionstr~te how it works, we ret)ort the results of our t)arsing exl)eriments on a Japanese tree bank.</Paragraph>
  </Section>
class="xml-element"></Paper>