{"updated":"2025-01-19T10:44:26.705241+00:00","links":{},"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00231497","sets":["1164:3500:11130:11413"]},"path":["11413"],"owner":"44499","recid":"231497","title":["マスク言語モデルによる日本語文章の自動強調付与"],"pubdate":{"attribute_name":"公開日","attribute_value":"2023-12-13"},"_buckets":{"deposit":"c8d4ccad-898d-4df3-b18b-c227223dedbf"},"_deposit":{"id":"231497","pid":{"type":"depid","value":"231497","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"マスク言語モデルによる日本語文章の自動強調付与","author_link":["625063","625062","625061"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"マスク言語モデルによる日本語文章の自動強調付与"},{"subitem_title":"Automatic Japanese Document Emphasis using Pre-Trained Masked Language Models","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"マスク言語モデル","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2023-12-13","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"現在，鳥取大学"},{"subitem_text_value":"現在，リコーITソリューションズ株式会社"},{"subitem_text_value":"現在，鳥取大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Presently with Tottori Uniersity","subitem_text_language":"en"},{"subitem_text_value":"Presently with Ricoh IT Solutions Co.,Ltd.","subitem_text_language":"en"},{"subitem_text_value":"Presently with Tottori Uniersity","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/231497/files/IPSJ-IFAT23153007.pdf","label":"IPSJ-IFAT23153007.pdf"},"date":[{"dateType":"Available","dateValue":"2025-12-13"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-IFAT23153007.pdf","filesize":[{"value":"1.1 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":"39"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"b8da4b26-a862-465b-8da8-23613739b9e3","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2023 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Zhuo, Binggang"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"本田, 涼太"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"村田, 真樹"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10114171","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_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2188-8884","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"本研究では，事前学習済みマスク言語モデルの RoBERTa を用いて，日本語文章の自動強調付与を行う手法を提案する．タイトル情報と条件付きランダムフィールドを組み合わせることで，提案手法は従来の研究とベースラインよりも大幅に性能が向上した．実験では，提案手法は平均 F1 スコア 0.437 を達成し，最も高性能だったベースライン（スコア 0.399）より 0.038 ，条件付きランダムフィールドを用いた従来研究（スコア 0.313）より 0.124 性能が向上した．本論文は広範囲の実験を通して，日本語文章の自動強調付与というタスクにおける様々な手法の性能を示した．有意水準 0.05 で両側ウィルコクソンの符号付き順位検定の結果は，提案手法がすべての比較手法と比較して統計的に有意であることを示した．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"This study proposes a method for automatic Japanese document emphasis using RoBERTa, a pre-trained mask language model. By combining title information and conditional random fields, our approach significantly improves performance over previous studies and carefully designed baselines. In the experiments, our method achieves an average F1-score of 0.437, improving by 0.038 over the best-performing baseline (score of 0.399), and by 0.124 over the previous study that used conditional random fields (score of 0.313). Through extensive experiments, we showcase the performance of various methods in the task of automatic Japanese document emphasis. The results of the two-sided Wilcoxon signed-rank test at a significance level of 0.05 indicated that the proposed method is statistically significant compared to all the comparison methods.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"6","bibliographic_titles":[{"bibliographic_title":"研究報告情報基礎とアクセス技術（IFAT）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2023-12-13","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"7","bibliographicVolumeNumber":"2023-IFAT-153"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":231497,"created":"2025-01-19T01:31:50.177798+00:00"}