<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <title>DSpace コミュニティ:</title>
  <link rel="alternate" href="http://hdl.handle.net/2065/6" />
  <subtitle />
  <id>http://hdl.handle.net/2065/6</id>
  <updated>2013-05-19T13:09:37Z</updated>
  <dc:date>2013-05-19T13:09:37Z</dc:date>
  <entry>
    <title>Capturing and Parsing the Mixed Properties of Light Verb Constructions in a Typed Feature Structure Grammar</title>
    <link rel="alternate" href="http://hdl.handle.net/2065/562" />
    <author>
      <name>Kim, Jong-Bok</name>
    </author>
    <author>
      <name>Yang, Jaehyung</name>
    </author>
    <author>
      <name>Choi, Incheol</name>
    </author>
    <id>http://hdl.handle.net/2065/562</id>
    <updated>2012-05-12T06:37:04Z</updated>
    <published>2005-11-16T07:15:16Z</published>
    <summary type="text">Title: Capturing and Parsing the Mixed Properties of Light Verb Constructions in a Typed Feature Structure Grammar
Authors: Kim, Jong-Bok; Yang, Jaehyung; Choi, Incheol
Abstract: One of the most widely used constructions in Korean is the so-called light verb construction (LVC) involving an active-denoting verbal noun (VN) together with the light verb ha-ta ‘do’. This paper first discusses the argument composition of the LVC, mixed properties of VNs which have provided a challenge to syntactic analyses with a strict version of X-bar theory. The paper shows the mechanism of multiple classification of category types with systematic inheritance can provide an effective way of capturing these mixed properties. The paper then restates the argument composition properties of the LVC and reenforces them with a constraint-based analysis. This paper also offers answers to the the puzzling syntactic variations in the LVC. Following these empirical and theoretical discussions is a short report on the implementation of the analysis within the LKB (Linguistics Knowledge Building) system.</summary>
    <dc:date>2005-11-16T07:15:16Z</dc:date>
  </entry>
  <entry>
    <title>Tiny Corpus Applications with Transformation-Based Error-Driven Learning : Evaluations of Automatic Grammar Induction and Partial Parsing of SaiSiyat</title>
    <link rel="alternate" href="http://hdl.handle.net/2065/573" />
    <author>
      <name>Lin, Zhemin</name>
    </author>
    <author>
      <name>Sung, Li-May</name>
    </author>
    <id>http://hdl.handle.net/2065/573</id>
    <updated>2012-05-12T07:20:18Z</updated>
    <published>2005-11-16T07:15:20Z</published>
    <summary type="text">Title: Tiny Corpus Applications with Transformation-Based Error-Driven Learning : Evaluations of Automatic Grammar Induction and Partial Parsing of SaiSiyat
Authors: Lin, Zhemin; Sung, Li-May
Abstract: This paper reports a preliminary result on automatic grammar induction based on the framework of Brill and Markus (1992) and binary-branching syntactic parsing of Esperanto and SaiSiyat (a Formosan language). Automatic grammar induction requires large corpus and is found implausible to process endangered minor languages. Syntactic parsing, on the contrary, needs merely tiny corpus and works along with corpora segmented by intonation-unit which results in high accuracy.</summary>
    <dc:date>2005-11-16T07:15:20Z</dc:date>
  </entry>
  <entry>
    <title>Japanese Subjects and Information Structure : A Constraint-based Approach</title>
    <link rel="alternate" href="http://hdl.handle.net/2065/563" />
    <author>
      <name>大谷 朗</name>
    </author>
    <author>
      <name>松本 裕治</name>
    </author>
    <id>http://hdl.handle.net/2065/563</id>
    <updated>2012-05-12T06:37:21Z</updated>
    <published>2005-11-16T07:15:16Z</published>
    <summary type="text">Title: Japanese Subjects and Information Structure : A Constraint-based Approach
Authors: 大谷 朗; 松本 裕治
Abstract: This paper is concerned with how topic/focus articulation should be optimally integrated into Japanese grammar. Based on Engdahl and Vallduv´.’s (1996) Information Structure, we propose an analysis with the following characteristics: (i) information structure is an integral part of Japanese grammar and interacts in principled ways with both syntax and phonology, (ii) the representations of topic/focus in the information structure and its interactions with the particles wa/ga show one-to-many relation, and (iii) the ordering of grammatical functions and its interactions with other grammatical parts play an important role in determining the focus domain.</summary>
    <dc:date>2005-11-16T07:15:16Z</dc:date>
  </entry>
  <entry>
    <title>On Argument-Adjunct Asymmetry of Sluicing in Mandarin Chinese</title>
    <link rel="alternate" href="http://hdl.handle.net/2065/576" />
    <author>
      <name>Chen, Li-Chi Lee</name>
    </author>
    <id>http://hdl.handle.net/2065/576</id>
    <updated>2012-05-12T06:36:11Z</updated>
    <published>2005-11-16T07:15:22Z</published>
    <summary type="text">Title: On Argument-Adjunct Asymmetry of Sluicing in Mandarin Chinese
Authors: Chen, Li-Chi Lee
Abstract: This study attempts to account for the argument-adjunct asymmetry of Sluicing in Mandarin Chinese. Such an asymmetry is empirically demonstrated by a language-particular phenomenon, so-called shi-support, which is also the last resort (Chomsky, 1995a) of our linguistic mechanism. In the current related literature, shi-support is obligatory for wh-arguments but optional for wh-adjuncts (Wang, 2002). However, I argue that at the PF level shi-support is even optional for wh-arguments; that is, it is only needed in the derivation at the LF level. My analysis is crucially based on CLM’s (1995) insightful analysis of LF Copying Theory. Departing from their analysis in crucial respects, however, I argue that a covert wh-movement also takes place simultaneously with the operation of copying the antecedent IP. For reasons of economy, such a non-overt movement is preferred and is of the least efforts (Procrastinate). In addition, evidence from shi-support argues that Sluicing in Mandarin Chinese prefers LF copying rather than PF deletion. To sum up, shi-support is compulsory for wh-arguments in that the ECP (Empty Category Principle) requirement must be satisfied at the LF level owning to the intervening “barriers” (Chomsky, 1986).</summary>
    <dc:date>2005-11-16T07:15:22Z</dc:date>
  </entry>
</feed>

