{"created":"2025-01-19T01:39:39.971424+00:00","updated":"2025-01-19T08:58:06.185327+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00237181","sets":["1164:6389:11481:11662"]},"path":["11662"],"owner":"44499","recid":"237181","title":["計算コスト削減を目的とした属性選択を用いるk-匿名化手法を活用した深層学習モデルの精度と安全性評価"],"pubdate":{"attribute_name":"公開日","attribute_value":"2024-07-15"},"_buckets":{"deposit":"3de9df2d-f0fb-423b-86c4-1071f6f2285f"},"_deposit":{"id":"237181","pid":{"type":"depid","value":"237181","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"計算コスト削減を目的とした属性選択を用いるk-匿名化手法を活用した深層学習モデルの精度と安全性評価","author_link":["648773","648775","648771","648774","648776","648772"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"計算コスト削減を目的とした属性選択を用いるk-匿名化手法を活用した深層学習モデルの精度と安全性評価"},{"subitem_title":"Evaluation of Accuracy and Security on Deep Learning Model by Utilizing K-Anonymization With Attribute Selection for Reducing Computational Cost","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"ISEC ","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2024-07-15","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"立命館大学理工学研究科"},{"subitem_text_value":"立命館大学理工学部"},{"subitem_text_value":"立命館大学理工学部"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Science and Engineering, Ritsumeikan University","subitem_text_language":"en"},{"subitem_text_value":"Department of Science and Engineering, Ritsumeikan University","subitem_text_language":"en"},{"subitem_text_value":"Department of Science and Engineering, Ritsumeikan 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/237181/files/IPSJ-SPT24056060.pdf","label":"IPSJ-SPT24056060.pdf"},"date":[{"dateType":"Available","dateValue":"2026-07-15"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SPT24056060.pdf","filesize":[{"value":"1.9 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"46"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"64a29e27-232a-49b8-bf67-8eb3316d7746","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2024 by the Institute of Electronics, Information and Communication Engineers This SIG report is only available to those in membership of the SIG."}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"上田, 澪"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"吉田, 康太"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"藤野, 毅"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Rei, Ueda","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Kota, Yoshida","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Takeshi, Fujino","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA12628305","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":"2188-8671","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"近年，パーソナルデータが様々な用途で活用されている．一般的にパーソナルデータには多くのプライバシー情報が含まれており，それらの収集や共有は困難である．これらのプライバシー情報を安全に共有するための技術として匿名化があり，k-匿名化はその一種である．k-匿名化はデータのプライバシーを保護できる一方で，その匿名化処理に多くの計算コストを要する課題がある．そこで本稿では k-匿名化の計算コストを削減することを目的に事前処理として属性選択を適用する．属性選択を適用した k-匿名化では，事前に選択した属性のみをデータとして残し，そのデータに対してのみ匿名化を行う．本稿では，属性選択を適用した k-匿名化データが深層学習モデルの学習データとして使用されることを想定し，有用性を深層学習モデルの精度として評価し，安全性を再識別攻撃とメンバーシップ推論攻撃で評価した．実験結果では，k-匿名化前に属性選択を適用しデータ量を削減することで，有用性や安全性を変化させずに匿名化の処理時間を短縮することができた．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"In recent years, personal data has been used for various purposes. In general, personal data contains a lot of private information, making it difficult to collect and share. Anonymization is a technique for securely sharing such private information, and k-anonymization is a typical method. Although k-anonymization can protect data privacy, the computational cost of k-anonymization is high. In this paper, we apply attribute selection as pre-processing to reduce the computational cost of k-anonymization. In this method, some attributes are deleted from the original data and k-anonymization is performed on the selected attributes. In this paper, we assumed that k-anonymized data is used to train a deep learning mode. Therefore, we evaluated the data usefulness by the trained model accuracy and privacy protection by re-identification attacks and membership inference attacks (MIAs). We showed that applying attribute selection to k-anonymization reduces the computational cost of anonymization without decreasing data utility and anonymity.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"8","bibliographic_titles":[{"bibliographic_title":"研究報告セキュリティ心理学とトラスト（SPT）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2024-07-15","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"60","bibliographicVolumeNumber":"2024-SPT-56"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":237181,"links":{}}