{"links":{},"id":56965,"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00056965","sets":["1164:5159:5180:5181"]},"path":["5181"],"owner":"1","recid":"56965","title":["係り受けとポーズ・フィラーの情報を用いた話し言葉の段階的チャンキング"],"pubdate":{"attribute_name":"公開日","attribute_value":"2005-12-22"},"_buckets":{"deposit":"064a5753-4bc7-4f50-b1ad-8015238f0acd"},"_deposit":{"id":"56965","pid":{"type":"depid","value":"56965","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"係り受けとポーズ・フィラーの情報を用いた話し言葉の段階的チャンキング","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"係り受けとポーズ・フィラーの情報を用いた話し言葉の段階的チャンキング"},{"subitem_title":"Cascaded Chunking of Spontaneous Japanese using Bunsetsu Dependency and Pause Information","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2005-12-22","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"京都大学情報学研究科知能情報学専攻"},{"subitem_text_value":" 京都大学学術情報メディアセンター"},{"subitem_text_value":" 京都大学学術情報メディアセンター  京都大学情報学研究科知能情報学専攻"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"School of Informatics, Kyoto University","subitem_text_language":"en"},{"subitem_text_value":" Academic Center for Computing and Media Studies, Kyoto University","subitem_text_language":"en"},{"subitem_text_value":" School of Informatics, Kyoto University, Academic Center for Computing and Media Studies, Kyoto University","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/56965/files/IPSJ-SLP05059048.pdf"},"date":[{"dateType":"Available","dateValue":"2007-12-22"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP05059048.pdf","filesize":[{"value":"981.4 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":"22"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"27d3ea97-89d1-410a-8199-0b9018091f64","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2005 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"西光, 雅弘"},{"creatorName":"高梨, 克也"},{"creatorName":"河原, 達也"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Masahiro, SAIKO","creatorNameLang":"en"},{"creatorName":"KatsuyaTAKANASHI","creatorNameLang":"en"},{"creatorName":"TatsuyaKAWAHARA","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10442647","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":"従来の音声言語処理においては，ポーズにより処理単位を設定することが大半である．しかし話し言葉では，ポーズにより認定した単位は文や節と必ずしも一致せず，均質な言語的まとまりになっているとは限らない．本研究では，話し言葉音声を村象として，頑健に抽出できる特徴として係り受けとポーズ・フィラーの情報に着目し，それらを段階的に用いることによりチャンキングを行う手法を提案する．まず，隣接文節間の係り受けの情報を用いて，構成要素にチャンキングする．この構成要素をポーズまたはフィラーの有無によってチャンキングすることにより，フレーズを生成する．『日本語話し言葉コーパス』（CSJ）で分析を行った結果，構成要素は主題や述語，格要素にあたる意味的なまとまりにおおむね対応しており，節境界の93％を検出できること，また，フレーズの過半数が節境界に対応し，特に絶対境界・強境界の94％を検出できることがわかった．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"This paper proposes a cascaded chunking method of spontaneous Japanese using bunsetsu dependency, pause and filler information. The proposed method generates two new units for spoken language processing: constituent and phrase. A constituent is generated by chunking bunsetsus using the adjacent bunsetsu dependency. A phrase is generated by chunking the constituents using pause and filler information. We made experiments using the Corpus of Spontaneous Japanese (CSJ). Most of constituents correspond to semantic chunks such as subjects, predicates and case frames, and 9397o of clause boundaries is detected. More than half of phrases correspond to the clause boundaries, and cover 94% of sentence boundaries.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"252","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"247","bibliographicIssueDates":{"bibliographicIssueDate":"2005-12-22","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"127(2005-SLP-059)","bibliographicVolumeNumber":"2005"}]},"relation_version_is_last":true,"weko_creator_id":"1"},"created":"2025-01-18T23:20:16.046878+00:00","updated":"2025-01-22T04:44:55.625830+00:00"}