{"created":"2025-01-19T00:21:57.963608+00:00","updated":"2025-01-20T17:56:09.119744+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00146510","sets":["1164:2735:7896:8400"]},"path":["8400"],"owner":"11","recid":"146510","title":["共分散行列の違いを許容する線形判別分析"],"pubdate":{"attribute_name":"公開日","attribute_value":"2015-12-08"},"_buckets":{"deposit":"50f4b4b5-816b-40aa-8a7f-2af39bc53222"},"_deposit":{"id":"146510","pid":{"type":"depid","value":"146510","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"共分散行列の違いを許容する線形判別分析","author_link":["228955","228956","228958","228957"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"共分散行列の違いを許容する線形判別分析"},{"subitem_title":"Novel Extended Fisher Discriminant Analysis","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2015-12-08","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"(株)NTTデータ"},{"subitem_text_value":"NTTコミュニケーション科学基礎研究所"},{"subitem_text_value":"福岡大学理学部応用数学科"},{"subitem_text_value":"福岡大学理学部応用数学科"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"NTT Data","subitem_text_language":"en"},{"subitem_text_value":"NTT Communication Science Lab.","subitem_text_language":"en"},{"subitem_text_value":"Fukuoka University","subitem_text_language":"en"},{"subitem_text_value":"Fukuoka 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/146510/files/IPSJ-MPS15106003.pdf","label":"IPSJ-MPS15106003.pdf"},"date":[{"dateType":"Available","dateValue":"2017-12-08"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-MPS15106003.pdf","filesize":[{"value":"196.7 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":"17"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"4aba462f-4a3e-4fed-9e43-2928be46aac7","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2015 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"坂野, 鋭"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"木村, 昭悟"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"藤木, 淳"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"田中, 勝"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10505667","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":"2188-8833","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"Fisher の判別分析 (FLDA) はその簡便さ，処理量の小ささ等から多くの研究者，分析実務者をひきつけてきた．しかしながら FLDA は，すべてのクラスの分布が同じ共分散構造を持っているという強い仮定があるために，この仮定が満たされない場合には識別精度の低下を引き起こすという問題点があった．我々は，データの分布を包摂する部分空間同士を引き離すという考え方に基づいて，この問題を解決する詳細化判別分析を提案する．提案手法は FLDA の適用範囲を広げるのみならず FLDA の次元数の制約を取り払う．また，処理量も大きくは無く，アルゴリズムも単純である．公開データによる実験的な評価の結果，提案手法の有効性が明らかになった．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Fisher's linear discriminant analysis (FLDA) has been attracting many researchers and practitioners for several decades thanks to its ease of use and low computational cost. However, FLDA implicitly assumes that all the classes share the same covariance, which implies that FLDA might fail when this assumption is not necessarily satisfied. To overcome this problem, we propose a simple extension of FLDA. The proposed method achieves remarkable improvements classification accuracy against FLDA while preserving two major strengths of FLDA: the ease of use and low computational costs. Experimental results with several data sets in UCI machine learning repository demonstrate the effectiveness of our method.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"5","bibliographic_titles":[{"bibliographic_title":"研究報告数理モデル化と問題解決（MPS）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2015-12-08","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"3","bibliographicVolumeNumber":"2015-MPS-106"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"id":146510,"links":{}}