World Wide Web を用いた外国人名の英訳自動獲得  [in Japanese] Automatic Acquisition of English Equivalents of Foreign Personal Names Using the World Wide Web  [in Japanese]

Access this Article

Search this Article

Author(s)

Abstract

カタカナで表記された外国人名の英訳語を,関連語をキーワードとする言語横断情報検索と,発音類似性を利用した訳語推定により,World Wide Web(WWW)文書から獲得する手法を提案する.ニュース記事に出現する人名は新出語であることが多く,対訳辞書に登録されていない場合も多い.提案手法は,カタカナの外国人名が文書中に存在した場合,はじめにその周辺の単語を対訳辞書によって英訳し,これらをキーワードとして英語のWWW 文書検索を行う.次に,検索されたWWW 文書中から人名候補となる英単語列を翻字により変換し,発音が類似した英単語列を訳語とする.ニュース記事に出現した外国人名を対象として本手法による実験を行い,有効性を確認した.This paper proposes a method of acquiring English equivalents of foreign personal names written in katakana characters from the World Wide Web (WWW). In news articles, new foreign personal names appear frequently and are rarely registered in bilingual dictionaries. Our method can automatically obtain the English equivalents of personal names by using two phases: cross-language information retrieval using related words and acquisition of translation based on phonetic similarity. In the first phase, given a katakana foreign personal name appearing in a news article, the method extracts words related to the foreign personal name, translates these words into English using bilingual dictionaries, and retrieves WWW documents in English using the translated words as keywords. In the second phase, our method extracts candidates of English equivalents from the retrievedWWWdocuments, transliterates the candidates to phonetic expressions, compares them with the phonetic expression of the personal name written in katakana, and obtains the most similar one as the English equivalent. We confirmed the effectiveness of our method with a series of experiments using foreign personal names appearing in news articles.

This paper proposes a method of acquiring English equivalents of foreign personal names written in katakana characters from the World Wide Web (WWW). In news articles, new foreign personal names appear frequently and are rarely registered in bilingual dictionaries. Our method can automatically obtain the English equivalents of personal names by using two phases: cross-language information retrieval using related words and acquisition of translation based on phonetic similarity. In the first phase, given a katakana foreign personal name appearing in a news article, the method extracts words related to the foreign personal name, translates these words into English using bilingual dictionaries, and retrieves WWW documents in English using the translated words as keywords. In the second phase, our method extracts candidates of English equivalents from the retrieved WWW documents, transliterates the candidates to phonetic expressions, compares them with the phonetic expression of the personal name written in katakana, and obtains the most similar one as the English equivalent. We confirmed the effectiveness of our method with a series of experiments using foreign personal names appearing in news articles.

Journal

  • IPSJ journal

    IPSJ journal 47(3), 968-979, 2006-03-15

    Information Processing Society of Japan (IPSJ)

References:  22

Codes

  • NII Article ID (NAID)
    110004708854
  • NII NACSIS-CAT ID (NCID)
    AN00116647
  • Text Lang
    JPN
  • Article Type
    Journal Article
  • ISSN
    1882-7764
  • NDL Article ID
    7875709
  • NDL Source Classification
    ZM13(科学技術--科学技術一般--データ処理・計算機)
  • NDL Call No.
    Z14-741
  • Data Source
    CJP  NDL  NII-ELS  IPSJ 
Page Top