{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00203656","sets":["1164:2735:10153:10154"]},"path":["10154"],"owner":"44499","recid":"203656","title":["河川水中病原体予測のための符号制約SVMと双対学習算法"],"pubdate":{"attribute_name":"公開日","attribute_value":"2020-02-24"},"_buckets":{"deposit":"d9831ea9-f7fb-4448-9307-d8398dcb7863"},"_deposit":{"id":"203656","pid":{"type":"depid","value":"203656","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"河川水中病原体予測のための符号制約SVMと双対学習算法","author_link":["502717","502716","502719","502718"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"河川水中病原体予測のための符号制約SVMと双対学習算法"}]},"item_type_id":"4","publish_date":"2020-02-24","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"群馬大学"},{"subitem_text_value":"群馬大学"},{"subitem_text_value":"東北大学"},{"subitem_text_value":"群馬大学／早稲田大学"}]},"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/203656/files/IPSJ-MPS20127020.pdf","label":"IPSJ-MPS20127020.pdf"},"date":[{"dateType":"Available","dateValue":"2022-02-24"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-MPS20127020.pdf","filesize":[{"value":"1.2 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":"17"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"8a6db105-17d5-4eda-a630-65affdc02c3d","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2020 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"土田, 康平"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"田島, 賢哉"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"佐野, 大輔"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"加藤, 毅"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10505667","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":"2188-8833","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"線形識別器の訓練において，ある特徴に正の相関があるというドメイン知識がある場合，訓練用データの個数が十分にあれば，対応する重み係数は正になることが好ましい．しかし，訓練用データ数が不十分，もしくは，データにノイズが多いとき，対応する重み係数が負に学習されてしまうことがある．我々は重み係数の符号を制約して線形識別器を学習する方法を考案し，河川水中病原体の予測への応用において有効性を確認してきた．本稿は，符号制約下で SVM を学習するための新しい最適化アルゴリズムを提案する．本研究で開発したアルゴリズムは，フランクウルフ法に基づいており，次の 3 点の長所を持つ：(i) 劣線形収束する; (ii )各反復の計算コストは O(nd); (iii) 停止条件が明確．すなわち，射影勾配法と同等の計算時間を持ちながら，さらに，反復を停止したときの解の精度を保証する算法となる．これらの理論保証は公開データセットを使った数値例で例示し，有効性を示す．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"6","bibliographic_titles":[{"bibliographic_title":"研究報告数理モデル化と問題解決（MPS）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2020-02-24","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"20","bibliographicVolumeNumber":"2020-MPS-127"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":203656,"updated":"2025-01-19T20:29:24.382890+00:00","links":{},"created":"2025-01-19T01:06:00.363204+00:00"}