{"created":"2025-01-18T23:35:13.338133+00:00","updated":"2025-01-21T19:36:18.110028+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00080867","sets":["1164:2735:6701:6702"]},"path":["6702"],"owner":"10","recid":"80867","title":["推薦システムのためのベイズ決定理論に基づくユニバーサルマルコフ決定過程"],"pubdate":{"attribute_name":"公開日","attribute_value":"2012-02-23"},"_buckets":{"deposit":"1af64d6b-1aa7-4cf1-b2d4-1e57f598dc9f"},"_deposit":{"id":"80867","pid":{"type":"depid","value":"80867","revision_id":0},"owners":[10],"status":"published","created_by":10},"item_title":"推薦システムのためのベイズ決定理論に基づくユニバーサルマルコフ決定過程","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"推薦システムのためのベイズ決定理論に基づくユニバーサルマルコフ決定過程"},{"subitem_title":"Universal Markov Decision Process Designed by Bayes Decision Theory for the Recommendation System","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2012-02-23","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"早稲田大学／株式会社NTTデータ"},{"subitem_text_value":"北見工業大学"},{"subitem_text_value":"早稲田大学"},{"subitem_text_value":"サイバー大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Waseda University / NTT DATA CORPORATION","subitem_text_language":"en"},{"subitem_text_value":"Kitami Institute of Technology","subitem_text_language":"en"},{"subitem_text_value":"Waseda University","subitem_text_language":"en"},{"subitem_text_value":"Cyber 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/80867/files/IPSJ-MPS12087010.pdf"},"date":[{"dateType":"Available","dateValue":"2014-02-23"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-MPS12087010.pdf","filesize":[{"value":"402.9 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":"17"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"b75b8916-3edf-4a17-a6b7-65471859a40c","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2012 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"桑田, 修平"},{"creatorName":"前田, 康成"},{"creatorName":"松嶋, 敏泰"},{"creatorName":"平澤, 茂一"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Shuhei, Kuwata","creatorNameLang":"en"},{"creatorName":"Yasunari, Maeda","creatorNameLang":"en"},{"creatorName":"Toshiyasu, Matsushima","creatorNameLang":"en"},{"creatorName":"Shigeichi, Hirasawa","creatorNameLang":"en"}],"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_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"推薦問題を扱うためのより一般化されたマルコフ決定過程モデルに対して，ベイズ基準のもとで最適な推薦ルールを履歴データから求める方法を提案する．提案法の特徴は，ある商品を推薦した後に何が買われたのかを考慮していること，さらに，一回の推薦結果だけでなく一定期間内に行った複数の推薦結果を評価している点にある．ここで，従来の推薦手法と大きく異なる点は，推薦ルールを求めるためのプロセスを統計的決定問題として厳密に定式化したことにある．その結果，推薦する目的に対して最適な推薦が行えるようになった．人工データを用いた評価実験により，提案する推薦手法の有効性を示す．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"In this paper, we proposed a general markov decision process model for the recommendation system. Furthermore, based on the bayesian decision theory, we derived the optimal recommendation lists from the proposed model using historical data. Our method takes into account not only the purchased items but also the past recommended items within a given period. Here, the unique thing about this paper is that we formulate the process to get the recommendation lists as the statistical decision problem. As a result, we can obtain the most suitable recommendation lists with respect to the purpose of the recommendation. We show the experimental results by using artificial data that our method can obtain more rewards than the conventional method gets.","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":"2012-02-23","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"10","bibliographicVolumeNumber":"2012-MPS-87"}]},"relation_version_is_last":true,"weko_creator_id":"10"},"id":80867,"links":{}}