{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00216008","sets":["1164:6757:10770:10771"]},"path":["10771"],"owner":"44499","recid":"216008","title":["機械学習を用いた画像識別におけるリスク分析手法の提案"],"pubdate":{"attribute_name":"公開日","attribute_value":"2022-01-13"},"_buckets":{"deposit":"23606659-3ee1-475b-acfd-ab956a4de7e4"},"_deposit":{"id":"216008","pid":{"type":"depid","value":"216008","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"機械学習を用いた画像識別におけるリスク分析手法の提案","author_link":["556908","556906","556909","556910","556911","556907"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"機械学習を用いた画像識別におけるリスク分析手法の提案"},{"subitem_title":"Proposal of Risk Analysis Method in Image Recognition Using Machine Learning","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"学習・分類","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2022-01-13","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"国立情報学研究所"},{"subitem_text_value":"慶應義塾大学大学院附属SDM研究所"},{"subitem_text_value":"国立情報学研究所"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"National Institute of Informatics","subitem_text_language":"en"},{"subitem_text_value":"Keio University SDM Research Institute","subitem_text_language":"en"},{"subitem_text_value":"National Institute of Informatics","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/216008/files/IPSJ-DCC22030028.pdf","label":"IPSJ-DCC22030028.pdf"},"date":[{"dateType":"Available","dateValue":"2024-01-13"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-DCC22030028.pdf","filesize":[{"value":"1.6 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":"50"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"5d1c8029-c28a-45ab-86a3-8ba5bd41f1ac","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2022 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"髙橋, 雄志"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"山口, 晋一"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"石川, 冬樹"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Yuji, Takahashi","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Shinichi, Yamaguchi","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Fuyuki, Ishikawa","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA12628338","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":"2188-8868","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"IoT 時代の到来により様々なシステムが連携した新しいサービスが登場するようなった.中でも人工知能などの機械学習を用いたシステムは,我々の生活に根差したサービスを提供するようになってきている.しかし,それらのサービスを構成するすべてのシステムにおける安全性が確保されているとは限らない.また,人に代わって自動判断を行うようなシステムでは機械学習の中でも画像識別を用いるものが多く,学習モデルの精度向上はサービスの品質向上につながる.これらのことから画像識別を用いた生活に根差したサービスを構築するにあたり,安全性を考慮した学習を行うことは重要であると考える.そこで本稿では,機械学習モデルのリスク分析を学習前と学習後それぞれの特徴に合わせて行うことを提案し,提案手法を実現するためのサポートツールの開発を行ったことを報告するものである.","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Technical phase changing to IoT, then new service was appeared that is several services cooperative. We have a service that uses machine learning that is closely related to our life. But, not all systems that make up those services are secure. In addition, many systems that make automatic decide to behalf of humans use image identification in machine learning and improving the accuracy of learning models leads to improving the quality of services. Based on the above, we consider that it is important to learning with safety in mind when constructing a service using image identification rooted in our life. We propose to perform risk analysis of machine learning models according to the characteristics before and after learning, and report that we have developed a support tool of the proposed method.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"6","bibliographic_titles":[{"bibliographic_title":"研究報告デジタルコンテンツクリエーション(DCC)"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2022-01-13","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"28","bibliographicVolumeNumber":"2022-DCC-30"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":216008,"updated":"2025-01-19T15:58:00.952364+00:00","links":{},"created":"2025-01-19T01:16:44.362366+00:00"}