{"links":{},"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00048244","sets":["1164:4179:4215:4216"]},"path":["4216"],"owner":"11","recid":"48244","title":["偽似訓練データを用いた語義曖昧性解消の改善"],"pubdate":{"attribute_name":"公開日","attribute_value":"2003-11-06"},"_buckets":{"deposit":"49d464c3-94c0-4cf7-891b-2b2cc1640c0b"},"_deposit":{"id":"48244","pid":{"type":"depid","value":"48244","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"偽似訓練データを用いた語義曖昧性解消の改善","author_link":["459080","459079","459081","459078"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"偽似訓練データを用いた語義曖昧性解消の改善"},{"subitem_title":"Improving Word Sense Disambiguation by Pseudo Training Data","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2003-11-06","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"奈良先端科学技術大学院大学情報科学研究科"},{"subitem_text_value":"奈良先端科学技術大学院大学情報科学研究科"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Information Science, Nara Institute of Science and Technology","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science, Nara Institute of Science and Technology","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/48244/files/IPSJ-NL03158018.pdf","label":"IPSJ-NL03158018"},"date":[{"dateType":"Available","dateValue":"2005-11-06"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-NL03158018.pdf","filesize":[{"value":"1.0 MB"}],"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":"191fa392-06f2-44eb-a76e-9c25dafb366c","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2003 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"王, 小捷"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"松本, 裕治"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Wang, Xiaojie","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Matsumoto, Yuji","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":"この論文は、語義曖昧性解消性脳を改善する方法を提案する。教師付分類器の特徴スペースを、語義が付けられたサンプルおよび偽似訓練データの両方からの特徴を組み合わせることにより拡張する。偽似訓練データを自動的に集めることができる方法を提案し，特徴スペースが十分に大きな規模まで拡張することができることを示す。また、語義が付けられたサンプル中に見えない単語の確率は偽似訓練データ中のその頻度によって平滑されるので、見えない特徴を区別することが出来る。実験は、この方法によってもたらされた著しい改善を示した。","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"This paper suggests a method to improve quality for word sense disambiguation. Feature spaces of supervised classifiers are extended and enhanced by combining features both from tagged samples and pseudo samples. Since pseudo samples can be collected automatically, feature spaces can be extended to a scale large enough, and probabilities for unseen words in tagged samples are smoothed according to its frequency in pseudo samples, which enable distinguishing between features unseen in tagged samples. Experiments have shown a significant improvement brought by this method.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"120","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告自然言語処理（NL）"}],"bibliographicPageStart":"115","bibliographicIssueDates":{"bibliographicIssueDate":"2003-11-06","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"108(2003-NL-158)","bibliographicVolumeNumber":"2003"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"updated":"2025-01-19T23:34:29.564124+00:00","created":"2025-01-18T23:13:32.288689+00:00","id":48244}