{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00047824","sets":["1164:4179:4187:4191"]},"path":["4191"],"owner":"1","recid":"47824","title":["NMF とリンクベースの修正法によるピンポン型文書クラスタリング"],"pubdate":{"attribute_name":"公開日","attribute_value":"2007-05-24"},"_buckets":{"deposit":"6ee65b1a-c17d-4026-afe5-65121ee079bd"},"_deposit":{"id":"47824","pid":{"type":"depid","value":"47824","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"NMF とリンクベースの修正法によるピンポン型文書クラスタリング","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"NMF とリンクベースの修正法によるピンポン型文書クラスタリング"},{"subitem_title":"Ping-Pong Document Clustering by using NMF and Linkage Based Refinement ","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2007-05-24","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"茨城大学工学部情報工学科"},{"subitem_text_value":"茨城大学工学部情報工学科"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Department of Computer and Information Sciences, Ibaraki University","subitem_text_language":"en"},{"subitem_text_value":"Department of Computer and Information Sciences, Ibaraki 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/47824/files/IPSJ-NL07179002.pdf"},"date":[{"dateType":"Available","dateValue":"2009-05-24"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-NL07179002.pdf","filesize":[{"value":"670.9 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":"23"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"1fc051e6-d0e4-4627-bab2-6aa3a4a28b01","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2007 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"新納, 浩幸"},{"creatorName":"佐々木, 稔"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Hiroyuki, Shinnou","creatorNameLang":"en"},{"creatorName":"Minoru, Sasaki","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10115061","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_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"Non-negative Matrix Factorization (NMF) は効果的な文書クラスタリング手法である。本論文では NMF の精度をさらに高めるために、NMF とリンクベースの修正法を交互に適用するピンポン型文書クラスタリング手法を提案する。NMF をピンポン型で利用することで、効果的な文書クラスタリングが期待できるが、NMF は入力のクラスタリング結果を改善できない場合も多く、ピンポン型で利用すると悪影響も多い。ここでは、ピンポンの終了条件の設定でこの問題に対処する。具体的には、リンクベースの修正法の適用の後に、設定した評価関数の値が改善されたかどうかでピンポンの終了を判定する。もし改善されていなければ、ピンポンを終了し、前回のリンクベースの修正法を行った後の結果を最終的なクラスタリング結果とする。これによって、NMF を利用した効果的なピンポン型クラスタリングが可能となる。実験では 16個の文書データセットを利用して、本手法を k-means や NMF と比較した。基本となる NMF の結果を大きく改善できた。","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Non-negative Matrix Factorization (NMF) is a powerful document clustering method.  This paper proposes a ping-pong document clustering method using NMF and the linkage based refinement alternately, in order to improve the clustering result of NMF. The use of NMF in the ping-pong strategy can be expected effective for document clustering.  However, NMF in the ping-pong strategy often worsens performance because NMF often fails to improve the clustering result given as the initial values.  Our method handles this problem with the stop condition of the ping-pong process.  Concretely speaking, our method decides the stop/continue of the ping-pong process by the value of an object function for the clustering result produced by the linkage based refinement.  If that value is not improved, our method stops the ping-pong process, and outputs the clustering result produced by the linkage based refinement in the previous ping-pong process.  By this setting, our method can use NMF in the ping-pong strategy.  In the experiment, we compared our method with the k-means and NMF by using 16 document data sets.  Our method improved the clustering result of NMF significantly.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"12","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告自然言語処理(NL)"}],"bibliographicPageStart":"7","bibliographicIssueDates":{"bibliographicIssueDate":"2007-05-24","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"47(2007-NL-179)","bibliographicVolumeNumber":"2007"}]},"relation_version_is_last":true,"weko_creator_id":"1"},"id":47824,"updated":"2025-01-22T08:48:14.902373+00:00","links":{},"created":"2025-01-18T23:13:12.696079+00:00"}