{"id":146843,"updated":"2025-01-20T17:49:52.033706+00:00","links":{},"created":"2025-01-19T00:22:10.738072+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00146843","sets":["6164:6165:6462:8443"]},"path":["8443"],"owner":"11","recid":"146843","title":["最小二乗密度比推定における差分プライバシー"],"pubdate":{"attribute_name":"公開日","attribute_value":"2015-10-14"},"_buckets":{"deposit":"6204feb4-6a9f-4bdb-b001-006836756c48"},"_deposit":{"id":"146843","pid":{"type":"depid","value":"146843","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"最小二乗密度比推定における差分プライバシー","author_link":["230756","230753","230754","230755","230758","230757"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"最小二乗密度比推定における差分プライバシー"},{"subitem_title":"Differentially Private Least-Squares Importance Fitting","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"PWS，差分プライバシー，PPDP，密度比推定，uLSIF","subitem_subject_scheme":"Other"}]},"item_type_id":"18","publish_date":"2015-10-14","item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_18_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"東京大学大学院情報理工学系研究科"},{"subitem_text_value":"東京大学情報基盤センター"},{"subitem_text_value":"東京大学情報基盤センター"}]},"item_18_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Dep. of Mathematical Informatics, The University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"Information Technology Center, The University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"Information Technology Center, The University of Tokyo","subitem_text_language":"en"}]},"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/146843/files/IPSJ-CSS2015051.pdf","label":"IPSJ-CSS2015051.pdf"},"date":[{"dateType":"Available","dateValue":"2017-10-14"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-CSS2015051.pdf","filesize":[{"value":"288.9 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":"30"},{"tax":["include_tax"],"price":"0","billingrole":"46"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"2db93049-9c1c-4b4e-a14d-fb8e585f6fc9","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2015 by the Information Processing Society of Japan"}]},"item_18_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"高林, 裕太"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"荒井, ひろみ"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"中川, 裕志"}],"nameIdentifiers":[{}]}]},"item_18_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Yuta, Takabayashi","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Hiromi, Arai","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Hiroshi, Nakagawa","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_5794","resourcetype":"conference paper"}]},"item_18_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"プライバシー保護データマイニングにおいて，非公開データベースに対する学習を既に公開されているデータベースで近似するというのは，現実的なシナリオである．このシナリオに対して差分プライバシーと呼ばれる保護概念からアプローチした Importance weighting mechanism は，確率的分類法と呼ばれる密度比推定手法に差分プライバシーを適用した手法である．一方，密度比推定自体の手法としては，より計算効率や精度の良い手法として，最小二乗密度比適合法；uLSIF が既に知られている．本稿では，この uLSIF に対して差分プライバシーを適用した手法を提案する．","subitem_description_type":"Other"}]},"item_18_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"In the privacy-preserving data mining, it is a possible scenario to approximate learning on a private database with a published database. The importance weighting mechanism, which approach the scenario with the privacy notion called Differential Privacy, is a differentially private extension of the importance estimation method called the Probabilistic Classification method. Meanwhile, another importance estimation method, the Least-squares Importance Fitting method called uLSIF, is already known as a more computationally efficient and accurate method. This paper proposes a differentially private uLSIF.","subitem_description_type":"Other"}]},"item_18_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"378","bibliographic_titles":[{"bibliographic_title":"コンピュータセキュリティシンポジウム2015論文集"}],"bibliographicPageStart":"371","bibliographicIssueDates":{"bibliographicIssueDate":"2015-10-14","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"3","bibliographicVolumeNumber":"2015"}]},"relation_version_is_last":true,"weko_creator_id":"11"}}