{"id":2006940,"created":"2026-02-03T05:47:07.856297+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:02006940","sets":["1164:1867:1770094923682:1770094831086"]},"path":["1770094831086"],"owner":"80578","recid":"2006940","title":["機械学習を用いたサーバーログ異常検知モデルの構築と評価"],"pubdate":{"attribute_name":"PubDate","attribute_value":"2026-02-09"},"_buckets":{"deposit":"0fec7d1f-1b5e-437d-8e12-6ad6c4bba7fa"},"_deposit":{"id":"2006940","pid":{"type":"depid","value":"2006940","revision_id":0},"owners":[80578],"status":"published","created_by":80578},"item_title":"機械学習を用いたサーバーログ異常検知モデルの構築と評価","author_link":[],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"機械学習を用いたサーバーログ異常検知モデルの構築と評価","subitem_title_language":"ja"},{"subitem_title":"Development and Evaluation of a Server Log Anomaly Detection Model Using Machine Learning","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"セキュリティ","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2026-02-09","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"摂南大学理工学部電気電子工学科情報系コース"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Setsunan University, Faculty of Science and Engineering, Department of Electrical and Electronic Engineering, Information Course","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/2006940/files/IPSJ-OS26170017.pdf","label":"IPSJ-OS26170017.pdf"},"date":[{"dateType":"Available","dateValue":"2028-02-09"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-OS26170017.pdf","filesize":[{"value":"883.6 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":"11"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"781530c3-35b8-4f72-a9be-f09bcd845fe5","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2026 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"西尾,魁浬"}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Kairi Nishio","creatorNameLang":"en"}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10444176","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":"2188-8795","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"オンラインゲームやWebサービスなどで発生するサーバー障害は，サービス品質の低下を招く重大な課題である．しかし，人力による監視ではヒューマンエラーにより発見が遅れるリスクが高い．そこで本研究では，サーバーログを用いた機械学習による高精度な異常検知モデルの構築を目的とする．具体的には，AWS EC2環境上で疑似的に障害を発生させ，CPU使用率などの特徴量を収集したデータをS3上に蓄積し，CSV化する．その後，Pandasを用いたデータ前処理を行い，ランダムフォレストやTensorFlowを用いた深層学習など，複数のアルゴリズムによるモデル構築と精度の比較を行う．評価実験を通じて，各手法の検知精度や特性を明らかにし，サーバー監視に最適なモデル構築を検討する．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Server failures in online games and Web services pose a significant challenge by degrading service quality. However, manual monitoring is prone to delayed detection caused by human error. Therefore, this study aims to construct a high-precision anomaly detection model using server logs and machine learning. Specifically, we simulate failures in an AWS EC2 environment, store collected feature data such as CPU usage in Amazon S3, and convert it to CSV format. After performing data preprocessing using Pandas, we construct and compare models using multiple algorithms, including Random Forest and Deep Learning with TensorFlow. Through evaluation experiments, we assess the detection accuracy and characteristics of each method to determine the optimal model for server monitoring.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"5","bibliographic_titles":[{"bibliographic_title":"研究報告システムソフトウェアとオペレーティング・システム（OS）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2026-02-09","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"17","bibliographicVolumeNumber":"2026-OS-170"}]},"relation_version_is_last":true,"weko_creator_id":"80578"},"updated":"2026-02-03T05:47:12.248303+00:00","links":{}}