<?xml version="1.0" standalone="yes"?>
<Paper uid="W02-1036">
  <Title>Combining Outputs of Multiple Japanese Named Entity Chunkers by Stacking</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> In the recent corpus-based NLP research, system combination techniques have been successfully applied to several tasks such as parts-of-speech tagging (van Halteren et al., 1998), base noun phrase chunking (Tjong Kim Sang, 2000), and parsing (Henderson and Brill, 1999; Henderson and Brill, 2000). The aim of system combination is to combine portions of the individual systems' outputs which are partial but can be regarded as highly accurate. The process of system combination can be decomposed into the following two sub-processes:  1. Collect systems which behave as differently as possible: it would help a lot if at least the col null lected systems tend to make errors of different types, because simple voting technique can identify correct outputs.</Paragraph>
    <Paragraph position="1"> Previously studied techniques for collecting such systems include: i) using several existing real systems (van Halteren et al., 1998; Brill and Wu, 1998; Henderson and Brill, 1999; Tjong Kim Sang, 2000), ii) bagging/boosting techniques (Henderson and Brill, 1999; Henderson and Brill, 2000), and iii) switching the data expression and obtaining several models (Tjong Kim Sang, 2000).</Paragraph>
  </Section>
class="xml-element"></Paper>