連体形形容詞に先行する格助詞「が」「の」格の係りに関する体系的分析  [in Japanese] Analysis on Dependency of GA/NO-particles preceding adjective in adnominal form  [in Japanese]

Access this Article

Search this Article

Author(s)

Abstract

長文の係り解析の精度向上は, 自然言語処理において重要な課題の一つである. 我々はすでに, 連体形形容詞周りの「が」「の」格に関して, 以下の3つのパターンに分類される7つの係りを規定するルールを見つけだした.<BR>・前後の名詞のみで係りが決まる.<BR>・前後の名詞と形容詞の関係で係りが決まる.<BR>・形容詞そのものの属性で係りが決まる.<BR>本論文では, 形容詞を網羅的に分析できるようにするために, 国立国語研究所での形容詞の体系的分類に従い分析対象形容詞を選択しその係りを調べた. それらの形容詞に対し7つのルールの妥当性を検証し拡張する. また, 対象形容詞を増やすことにより2つの新たなルールを検出することができた. これら, 分類を網羅するように選択した形容詞に対しても約95%の精度で係りを決めることができた.

Improving the accuracy of dependency analysis in long Japanese sentences is a big problem in the natural language analysis.Concerning the form of “‹noun-1›‹adjective›+‹noun-2›”, we found the seven effective dependency rules which are classified into three following patterns in the previous paper.<BR>·determined only from preceding or following nouns.<BR>·determined from the relation between preceding or following nouns and adjective.<BR>·determined from the characteristic of adjective itself.<BR>In this paper, we research more systematically the dependency of adnominal which contains I-adjective or NA-adjective, based on classification of adjectives. We explain (1) these seven rules are effectively applied to new adjectives by some extension, (2) some other rules are effective to improve the accuracy of dependency.Finally we obtained an accuracy of about 95% by applying our method to determine the dependency of adnominals which included an adjective.

Journal

  • Journal of Natural Language Processing

    Journal of Natural Language Processing 6(7), 109-123, 1999-10-10

    The Association for Natural Language Processing

References:  4

Codes

  • NII Article ID (NAID)
    10008829576
  • NII NACSIS-CAT ID (NCID)
    AN10472659
  • Text Lang
    JPN
  • Article Type
    ART
  • ISSN
    13407619
  • NDL Article ID
    4888796
  • NDL Call No.
    Z21-B168
  • Data Source
    CJP  NDL  J-STAGE 
Page Top