{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00213641","sets":["1164:3925:10503:10715"]},"path":["10715"],"owner":"44499","recid":"213641","title":["APIコール情報を用いた注意機構付きLSTMによるマルウェアの特徴抽出と分類"],"pubdate":{"attribute_name":"公開日","attribute_value":"2021-11-01"},"_buckets":{"deposit":"e2f6f862-17e4-402c-89db-df04c6fdfd3c"},"_deposit":{"id":"213641","pid":{"type":"depid","value":"213641","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"APIコール情報を用いた注意機構付きLSTMによるマルウェアの特徴抽出と分類","author_link":["546939","546938","546940","546941","546937","546942"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"APIコール情報を用いた注意機構付きLSTMによるマルウェアの特徴抽出と分類"},{"subitem_title":"Malware Classification and Feature Extraction Using Attention-Based LSTM and API Call Information","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2021-11-01","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"立命館大学"},{"subitem_text_value":"立命館大学"},{"subitem_text_value":"立命館大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Ritsumeikan University","subitem_text_language":"en"},{"subitem_text_value":"Ritsumeikan University","subitem_text_language":"en"},{"subitem_text_value":"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/213641/files/IPSJ-CSEC21095010.pdf","label":"IPSJ-CSEC21095010.pdf"},"date":[{"dateType":"Available","dateValue":"2023-11-01"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-CSEC21095010.pdf","filesize":[{"value":"340.5 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":"44"}],"accessrole":"open_date","version_id":"ba052c65-8319-42bc-9705-27c0bebe19b8","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2021 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"大江, 弘晃"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"毛利, 公一"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"鄭, 俊俊"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Hiroaki, Oe","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Koichi, Mouri","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Junjun, Zheng","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA11235941","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":"2188-8655","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"機械学習を用いた様々なマルウェア分類手法が提案されている．しかし，これらの分類手法のほとんどは，分類精度のみに着目したものであり，高い分類性能を実現するための効果的な特徴抽出についてはあまり着目されていない．そこで，本稿では画像やテキストの分類において，注目点の可視化に利用されるニューラルネットワーク技術の一つである注意機構（Attention mechanism）を用いて，マルウェア分類精度の向上に寄与する特徴抽出の可能性について述べる．具体的には，重要な特徴を明らかにし，それにより高精度な分類を実現する注意機構付き LSTM (Long Short-Term Memory) を提案する．本稿では，特に，Windows API の関数名および引数の時系列データを用いた分類を行い，注意機構の有無による分類精度の比較を行う．また，注意機構から抽出した重み値から重要度が高いと判断された特徴が，対象のファミリに由来する特徴かを確認する．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Various malware classification methods using machine learning have been proposed. However, Almost all the proposed classification methods focused on only the classification accuracy, and didn't pay much attention to the effective feature extraction achieving high classification performance. Therefore, in this paper, we attempt to identify important features contributing more to the malware classification accuracy using attention mechanism, which is one of the neural network technologies widely adopted to visualize the points of interest in image and text classification. Specifically, we propose an attention-based long short-term memory (LSTM) method to reveal important features and thereby achieve a high-accuracy classification. In this paper, in particular, we use the time series data of Windows API calls, containing both function names and arguments, and compare the classification accuracies with and without the attention mechanism. In addition, the weight values extracted from the attention mechanism confirm whether the features identified as being of high importance originate from the target family.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"8","bibliographic_titles":[{"bibliographic_title":"研究報告コンピュータセキュリティ（CSEC）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2021-11-01","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"10","bibliographicVolumeNumber":"2021-CSEC-95"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":213641,"updated":"2025-01-19T17:05:12.351129+00:00","links":{},"created":"2025-01-19T01:14:29.830170+00:00"}