{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00228865","sets":["934:989:11239:11352"]},"path":["11352"],"owner":"44499","recid":"228865","title":["メールフィルタリング性能向上のための未学習辞書なし語の効果"],"pubdate":{"attribute_name":"公開日","attribute_value":"2023-10-31"},"_buckets":{"deposit":"12607b7f-0939-4854-ae62-eda32933fae0"},"_deposit":{"id":"228865","pid":{"type":"depid","value":"228865","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"メールフィルタリング性能向上のための未学習辞書なし語の効果","author_link":["614414","614412","614415","614413"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"メールフィルタリング性能向上のための未学習辞書なし語の効果"},{"subitem_title":"Effect of Untrained Strange Words in Mail Filtering","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"[オリジナル論文] メールフィルタリング，辞書にない語，未学習語，出現傾向の変化","subitem_subject_scheme":"Other"}]},"item_type_id":"3","publish_date":"2023-10-31","item_3_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"山口大学大学院創成科学研究科／現在，福山大学工学部情報工学科"},{"subitem_text_value":"山口大学大学院創成科学研究科"}]},"item_3_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Sciences and Technology for Innovation, Yamaguchi University / Presently with Faculty of Engineering, Department of Computer Science, Fukuyama University","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Sciences and Technology for Innovation, Yamaguchi 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/228865/files/IPSJ-TOM1602013.pdf","label":"IPSJ-TOM1602013.pdf"},"date":[{"dateType":"Available","dateValue":"2025-10-31"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-TOM1602013.pdf","filesize":[{"value":"4.1 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":"17"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"92711425-4ef9-41a4-9f11-6bc8de03409e","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2023 by the Information Processing Society of Japan"}]},"item_3_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"天満, 誠也"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"松野, 浩嗣"}],"nameIdentifiers":[{}]}]},"item_3_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Seiya, Temma","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Hiroshi, Matsuno","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_3_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA11464803","subitem_source_identifier_type":"NCID"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_6501","resourcetype":"journal article"}]},"item_3_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"1882-7780","subitem_source_identifier_type":"ISSN"}]},"item_3_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"これまでに，機械学習に基づくメールフィルタリング手法が多く提案され，完全フィルタリングに近い分類が可能となっている．これらの手法は，過去に受信したメールから抽出した特徴に基づき，新規に受信したメールを分類するため，特徴が変化したメールに対する分類性能は低くなる．このようなメールを正しく分類するためには，新しく出現した機械学習していない単語からも特徴を抽出する必要がある．そこで，これらの出現傾向について調べたところ，時間経過とともに，スパムメールに辞書に載っていない単語が多くなる傾向にあることが分かった．この特徴を分類に利用するため，未学習の辞書に載っていない単語が多いメールを，スパムメールに分類しやすくする手法を提案した．この手法の適用効果を調べた結果，未学習の辞書に載っていない単語に，0.7付近というある程度高いスパム確率を設定することで，フィルタリング性能が向上することが分かった．","subitem_description_type":"Other"}]},"item_3_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Many machine learning-based mail filtering methods, so far proposed, have high performance, closing to perfect filtering. These methods, however, usually decrease performance for mails with unstable features, because they classify mails based on features of older mails. In order to classify such mails, it is necessary to extract features from new words which are not processed in machine learning. By examining appearance tendency of strange words (words which are not in dictionary), we found that the number of these words increase over time after a process of machine learning. We conducted an experiment classifying untrained strange words with varying spam probability, and found that more high performance classification is possible by setting 0.7 as a spam probability.","subitem_description_type":"Other"}]},"item_3_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"127","bibliographic_titles":[{"bibliographic_title":"情報処理学会論文誌数理モデル化と応用（TOM）"}],"bibliographicPageStart":"118","bibliographicIssueDates":{"bibliographicIssueDate":"2023-10-31","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"2","bibliographicVolumeNumber":"16"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":228865,"updated":"2025-01-19T11:41:26.143609+00:00","links":{},"created":"2025-01-19T01:27:59.949454+00:00"}