{"id":2005105,"created":"2025-10-23T04:30:50.916891+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:02005105","sets":["6164:6165:6640:1752804461720"]},"path":["1752804461720"],"owner":"11","recid":"2005105","title":["訓練データへの実行可能なノイズ付与によるIoTマルウェア画像分類手法の堅牢性向上の検証"],"pubdate":{"attribute_name":"PubDate","attribute_value":"2025-06-18"},"_buckets":{"deposit":"fe140973-7d99-4457-bb94-25bcb3372b9a"},"_deposit":{"id":"2005105","pid":{"type":"depid","value":"2005105","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"訓練データへの実行可能なノイズ付与によるIoTマルウェア画像分類手法の堅牢性向上の検証","author_link":[],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"訓練データへの実行可能なノイズ付与によるIoTマルウェア画像分類手法の堅牢性向上の検証","subitem_title_language":"ja"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"ネットワーク・システムセキュリティ(1)","subitem_subject_scheme":"Other"}]},"item_type_id":"18","publish_date":"2025-06-18","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":"はこだて未来大"},{"subitem_text_value":"はこだて未来大"},{"subitem_text_value":"はこだて未来大"}]},"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/2005105/files/IPSJ-DICOMO2025080.pdf","label":"IPSJ-DICOMO2025080.pdf"},"date":[{"dateType":"Available","dateValue":"2027-06-18"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-DICOMO2025080.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":"44"}],"accessrole":"open_date","version_id":"7c6534ad-901a-48bf-883e-6d6f8a92833a","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2025 by the Information Processing Society of Japan"}]},"item_18_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"川田, 隼大"}]},{"creatorNames":[{"creatorName":"稲村, 浩"}]},{"creatorNames":[{"creatorName":"石田, 繁巳"}]}]},"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":"マルウェアをグレースケール画像に変換し，CNN を用いて分類する手法が注目されている．このような分類手法は軽量かつ高い精度を示す一方，画像へノイズを加えることで誤分類を引き起こす脆弱性を持つ．本研究では，画像へのノイズ追加に耐性を持たせるため，分類精度と誤分類検体の予測確信度を目的関数とした訓練データの構成比最適化手法を提案する．提案手法の有効性を検証するため，ベースライン画像分類器，単一ノイズ種で構成された画像分類器，最適構成の画像分類器の三種を用いて比較実験を行った．その結果，訓練データへのノイズ追加によって分類精度が改善され，さらに構成比を最適化することで，誤分類検体における予測確信度が抑制され，堅牢性向上に有効なことが示された．","subitem_description_type":"Other"}]},"item_18_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"599","bibliographic_titles":[{"bibliographic_title":"マルチメディア，分散，協調とモバイル(DICOMO2025)シンポジウム2025論文集"}],"bibliographicPageStart":"592","bibliographicIssueDates":{"bibliographicIssueDate":"2025-06-18","bibliographicIssueDateType":"Issued"},"bibliographicVolumeNumber":"2025"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"updated":"2025-10-23T04:50:58.568225+00:00","links":{}}