{"created":"2021-03-01T06:20:34.715125+00:00","id":13543,"links":{},"metadata":{"_buckets":{"deposit":"94d25dc9-d2a7-41f9-bcfc-01d79e617eea"},"_deposit":{"id":"13543","owners":[],"pid":{"revision_id":0,"type":"depid","value":"13543"},"status":"published"},"_oai":{"id":"oai:nagoya.repo.nii.ac.jp:00013543","sets":["320:502:670"]},"author_link":["42743","42744","42745"],"item_10_alternative_title_19":{"attribute_name":"その他のタイトル","attribute_value_mlt":[{"subitem_alternative_title":"Effects of Number of Subjects on Activity Recognition : HASC2010corpus","subitem_alternative_title_language":"en"}]},"item_10_biblio_info_6":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicIssueDates":{"bibliographicIssueDate":"2011-07-06","bibliographicIssueDateType":"Issued"},"bibliographicPageEnd":"82","bibliographicPageStart":"76","bibliographic_titles":[{"bibliographic_title":"マルチメディア、分散、協調とモバイル(DICOMO)シンポジウム論文集","bibliographic_titleLang":"ja"}]}]},"item_10_description_4":{"attribute_name":"抄録","attribute_value_mlt":[{"subitem_description":"近年，人間行動認識に関する研究は多くの研究者によって取り組まれている．これらの研究の多くでは，各研究で提案された行動認識アルゴリズムを評価する際に，少数の被験者のみを対象とした実験を行っており，評価用データの大きさが十分であるとはいえない．また，各々の研究室で被験者から収集した行動データや被験者情報は公開されていない場合が多く，これらの情報が研究室間で共有されることはほとんどない．上記の問題を解決するために，我々は，500 名程度の被験者情報，6000 以上の行動データファイルを含む人間行動データコーパス”HASC2010corpus ”を構築した．本論文では，被験者数と人間行動認識率の相関に関する知見を得ることを目的とし，HASC2010corpus を対象として，被験者数を変化させた場合の認識率について評価を行った．評価の際には，行動認識分野の多くの研究で用いられている基本的な認識手法を参照し，特徴量の数と被験者数を変化させながらユーザ依存／ユーザ独立認識を行った．実験結果から，被験者数が大規模になるにつれ認識率が向上することが確認された．特に，ユーザ独立認識を行った際の認識率の向上が顕著であった．被験者60 名を対象とし，決定木を用いた機械学習手法を適用したユーザ依存認識を行ったところ，73%程度の認識率が得られた．尚，本論文ではのべ13000 回を超える評価評価実験を行った．評価実験の結果から，特にユーザ独立認識において，大規模行動データが重要であることを確認することができた．また，より高度な認識アルゴリズムとより大規模な被験者データを用いれば，実環境における行動認識率の向上が期待されることが分かった．","subitem_description_language":"ja","subitem_description_type":"Abstract"},{"subitem_description":"In recent years, there are many researches about the human activity recognition. In the most of these researches, the activity recognition experiments are performed with only a small number of subjects. Additionally, there is no public database of activity data or reference information about the suitable number of subjects. To overcome the situation, we have collected ”HASC2010corpus” with more than 6000 activity data from more than 500 test subjects. In this paper, we report the result of the experiments on the effects of the number of subjects as a basic reference for the field of activity recognition research. By using the part of HASC2010corpus, we performed a large number of evaluations with the user-dependent / user-independent data, different number of the features, and the different number of subjects. In the case of using user-dependent data, decision tree classifiers showed the activity recognition results with an overall accuracy rate of 73targeted 13869 description subjects. The result of experiments shows the importance of the large number of subjects especially on the user-independent data. The result also suggests that by using more advanced features and more large number of subjects, the activity recognition rate might be improved.","subitem_description_language":"en","subitem_description_type":"Abstract"}]},"item_10_identifier_60":{"attribute_name":"URI","attribute_value_mlt":[{"subitem_identifier_type":"HDL","subitem_identifier_uri":"http://hdl.handle.net/2237/15442"}]},"item_10_publisher_32":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"一般社団法人情報処理学会","subitem_publisher_language":"ja"}]},"item_10_rights_12":{"attribute_name":"権利","attribute_value_mlt":[{"subitem_rights":"ここに掲載した著作物の利用に関する注意 本著作物の著作権は（社）情報処理学会に帰属します。本著作物は著作権者である情報処理学会の許可のもとに掲載するものです。ご利用に当たっては「著作権法」ならびに「情報処理学会倫理綱領」に従うことをお願いいたします。 Notice for the use of this material The copyright of this material is retained by the Information Processing Society of Japan (IPSJ). This material is published on this web site with the agreement of the author (s) and the IPSJ. Please be complied with Copyright Law of Japan and the Code of Ethics of the IPSJ if any users wish to reproduce, make derivative work, distribute or make available to the public any part or whole thereof. All Rights Reserved, Copyright (C) Information Processing Society of Japan. Comments are welcome. Mail to address editj<at>ipsj.or.jp, please.","subitem_rights_language":"ja"}]},"item_10_select_15":{"attribute_name":"著者版フラグ","attribute_value_mlt":[{"subitem_select_item":"publisher"}]},"item_1615787544753":{"attribute_name":"出版タイプ","attribute_value_mlt":[{"subitem_version_resource":"http://purl.org/coar/version/c_970fb48d4fbd8a85","subitem_version_type":"VoR"}]},"item_access_right":{"attribute_name":"アクセス権","attribute_value_mlt":[{"subitem_access_right":"open access","subitem_access_right_uri":"http://purl.org/coar/access_right/c_abf2"}]},"item_creator":{"attribute_name":"著者","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"小川, 延宏","creatorNameLang":"ja"}],"nameIdentifiers":[{"nameIdentifier":"42743","nameIdentifierScheme":"WEKO"}]},{"creatorNames":[{"creatorName":"梶, 克彦","creatorNameLang":"ja"}],"nameIdentifiers":[{"nameIdentifier":"42744","nameIdentifierScheme":"WEKO"}]},{"creatorNames":[{"creatorName":"河口, 信夫","creatorNameLang":"ja"}],"nameIdentifiers":[{"nameIdentifier":"42745","nameIdentifierScheme":"WEKO"}]}]},"item_files":{"attribute_name":"ファイル情報","attribute_type":"file","attribute_value_mlt":[{"accessrole":"open_date","date":[{"dateType":"Available","dateValue":"2018-02-20"}],"displaytype":"detail","filename":"dt_hiro.pdf","filesize":[{"value":"325.1 kB"}],"format":"application/pdf","licensetype":"license_note","mimetype":"application/pdf","url":{"label":"dt_hiro.pdf","objectType":"fulltext","url":"https://nagoya.repo.nii.ac.jp/record/13543/files/dt_hiro.pdf"},"version_id":"2f590c8d-b081-4c6f-af34-d87f509e3895"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourcetype":"journal article","resourceuri":"http://purl.org/coar/resource_type/c_6501"}]},"item_title":"HASC2010corpus を用いた被験者数と人間行動認識率の相関分析","item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"HASC2010corpus を用いた被験者数と人間行動認識率の相関分析","subitem_title_language":"ja"}]},"item_type_id":"10","owner":"1","path":["670"],"pubdate":{"attribute_name":"PubDate","attribute_value":"2011-09-06"},"publish_date":"2011-09-06","publish_status":"0","recid":"13543","relation_version_is_last":true,"title":["HASC2010corpus を用いた被験者数と人間行動認識率の相関分析"],"weko_creator_id":"1","weko_shared_id":-1},"updated":"2023-01-16T04:00:47.539511+00:00"}