{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00048148","sets":["1164:4179:4208:4211"]},"path":["4211"],"owner":"1","recid":"48148","title":["慣用句抽出のための統計尺度の比較評価"],"pubdate":{"attribute_name":"公開日","attribute_value":"2004-07-16"},"_buckets":{"deposit":"f5bd1314-673c-455e-95c3-ff28354d2838"},"_deposit":{"id":"48148","pid":{"type":"depid","value":"48148","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"慣用句抽出のための統計尺度の比較評価","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"慣用句抽出のための統計尺度の比較評価"},{"subitem_title":"Extraction of Japanese noun - verb collocations from Corpora","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2004-07-16","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"（株）日立製作所中央研究所"},{"subitem_text_value":"（株）日立製作所中央研究所"},{"subitem_text_value":"（株）日立製作所中央研究所"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Central Research Laboratory, Hitachi, Ltd.","subitem_text_language":"en"},{"subitem_text_value":"Central Research Laboratory, Hitachi, Ltd.","subitem_text_language":"en"},{"subitem_text_value":"Central Research Laboratory, Hitachi, Ltd.","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_publisher":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"情報処理学会","subitem_publisher_language":"ja"}]},"publish_status":"0","weko_shared_id":-1,"item_file_price":{"attribute_name":"Billing file","attribute_type":"file","attribute_value_mlt":[{"url":{"url":"https://ipsj.ixsq.nii.ac.jp/record/48148/files/IPSJ-NL04162015.pdf"},"date":[{"dateType":"Available","dateValue":"2006-07-16"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-NL04162015.pdf","filesize":[{"value":"454.3 kB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"660","billingrole":"5"},{"tax":["include_tax"],"price":"330","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"23"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"ae4daab4-2943-4429-b6c0-4f2f6c286ab0","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2004 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"相薗, 敏子"},{"creatorName":"小泉, 敦子"},{"creatorName":"森本, 康嗣"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Toshiko, Aizono","creatorNameLang":"en"},{"creatorName":"Atsuko, Koizumi","creatorNameLang":"en"},{"creatorName":"Yasutsugu, Morimoto","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10115061","subitem_source_identifier_type":"NCID"}]},"item_4_textarea_12":{"attribute_name":"Notice","attribute_value_mlt":[{"subitem_textarea_value":"SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc."}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_18gh","resourcetype":"technical report"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"テキストマイニングでは，文書DBから単語や単語のペアを抽出して文書DBの特徴としてユーザに表示する。慣用句は，複数の単語で一つの意味を表すため，全体でまとまった単位として扱う必要がある。本研究では，名詞と動詞のペアからなる慣用句を文書DBから抽出する方式を検討した。本稿では，慣用句を抽出するための統計尺度の比較・評価について述べる。相互情報量，χ^2，AICの比較実験により，χ^2が統計尺度としては最適であることを確認した。また異分野の文書DBを利用することにより，F値が50％から53％に向上した。","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Extraction of noun-verb relations is essential for knowledge extraction from text.  Proper treatment of collocations is one of the tasks, since they often present non-constitutional concepts and have to be treated as single units instead of verb-noun relations.  In this paper, we describe an experimental evaluation of statistical measures for extracting Japanese noun-verb collocations from a corpus.  Based on the result of an experiment using three measures; Mutual Information, χ^2, as well as AIC, we conclude that χ^2 is the most suitable measure to extract collocations.  F measeure of χ^2 was improved from 50.1% to 52.8%　by using  a corpus from a different domain.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"108","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告自然言語処理（NL）"}],"bibliographicPageStart":"103","bibliographicIssueDates":{"bibliographicIssueDate":"2004-07-16","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"73(2004-NL-162)","bibliographicVolumeNumber":"2004"}]},"relation_version_is_last":true,"weko_creator_id":"1"},"id":48148,"updated":"2025-01-22T08:38:51.179750+00:00","links":{},"created":"2025-01-18T23:13:27.756259+00:00"}