Japanese Hyponymy Extraction based on a Term Similarity Graph

• Item type: SIG Technical Reports(1)
• 公開日: 2011-11-15
• タイトル: Japanese Hyponymy Extraction based on a Term Similarity Graph
• タイトル / 言語: en
• タイトル: Japanese Hyponymy Extraction based on a Term Similarity Graph
• 言語: eng
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: The University of Tokyo
• 著者所属: Microsoft Research Asia
• 著者所属(英): en
• 著者所属(英): The University of Tokyo
• 著者所属(英): en
• 著者所属(英): Microsoft Research Asia
• 著者名: Takuya, Akiba; Tetsuya, Sakai
• 著者名(英): Takuya, Akiba; Tetsuya, Sakai
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: We present a new method for automatic extraction of hyponymy relations between Japanese words from large-scale web corpora. Our method utilizes a term similarity graph, as well as information from Wikipedia. Our experimental results based on tens of millions of web pages show that our method can extract Japanese hyponymy relations with 82% precision.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: We present a new method for automatic extraction of hyponymy relations between Japanese words from large-scale web corpora. Our method utilizes a term similarity graph, as well as information from Wikipedia. Our experimental results based on tens of millions of web pages show that our method can extract Japanese hyponymy relations with 82% precision.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AN10114171
• 書誌情報: 研究報告情報基礎とアクセス技術（IFAT）; 巻 2011-IFAT-104, 号 3, p. 1-6, 発行日 2011-11-15
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会