<?xml version="1.0" standalone="yes"?>
<Paper uid="P05-2024">
  <Title>Corpus-Oriented Development of Japanese HPSG Parsers</Title>
  <Section position="3" start_page="0" end_page="139" type="intro">
    <SectionTitle>
2 HPSG
</SectionTitle>
    <Paragraph position="0"> Head-Driven Phrase Structure Grammar (HPSG) is classified into lexicalized grammars (Schabes et al., 1988). It attempts to model linguistic phenomena by interactions between a small number of grammar rules and a large number of lexical entries. Figure 1 shows an example of an HPSG derivation of a Japanese sentence 'kare ga shinda,' which means, 'He died.' In HPSG, linguistic entities such as words and phrases are represented by typed feature structures called signs, and the grammaticality of a sentence is verified by applying grammar rules to a sequence of signs. The sign of a lexical entry encodes the type and valence (i.e. restriction on the types of phrases that can appear around the word) of a corresponding word. Grammar rules of HPSG consist of  schemata and principles, the former enumerate possible patterns of phrase structures, and the latter are basically for controlling the inheritance of daughters' features to the parent.</Paragraph>
    <Paragraph position="1"> In the current example, the lexical entry for &amp;quot;shinda&amp;quot; is of the type verb, as indicated in its HEAD, and its COMPS feature restricts its preceding phrase to be of the type PP&amp;quot;ga&amp;quot;. The HEAD feature of the root node of the derivation is inherited from the lexical entry for &amp;quot;shinda&amp;quot;, because complement-head structures are head-final, and the head feature principle states that the HEAD feature of a phrase must be inherited from its head daughter.</Paragraph>
    <Paragraph position="2"> There are several implementations of Japanese HPSG grammars. JACY (Siegel and Bender, 2002) is a hand-crafted Japanese HPSG grammar that provides semantic information as well as linguistically motivated analysis of complex constructions. However, the evaluation of the grammar has not been done on domain-independent real-world texts such as newspaper articles. Although Bond et al. (2004) attempted to improve the coverage of the JACY grammar through the development of an HPSG treebank, they limited the target of their treebank annotation to short sentences from dictionary definitions. SLUNG (Mitsuishi et al., 1998) is an HPSG grammar whose coverage on real-world sentences is about 99%, but the grammar is underspecified, which means that the constraints of the grammar are not sufficient for conducting semantic analysis. By employing corpus-oriented development, we aim to develop a wide-coverage HPSG parser that enables</Paragraph>
  </Section>
class="xml-element"></Paper>