{"created":"2025-01-19T01:44:49.350645+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00240561","sets":["1164:3027:11450:11786"]},"path":["11786"],"owner":"44499","recid":"240561","title":["リザバーコンピューティングによる生体情報を用いた省資源かつ効率的な感情推定手法の提案"],"pubdate":{"attribute_name":"公開日","attribute_value":"2024-11-11"},"_buckets":{"deposit":"7cd228ef-af90-4dd9-b214-dfbd6efb026b"},"_deposit":{"id":"240561","pid":{"type":"depid","value":"240561","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"リザバーコンピューティングによる生体情報を用いた省資源かつ効率的な感情推定手法の提案","author_link":["660199","660202","660203","660205","660200","660206","660204","660201"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"リザバーコンピューティングによる生体情報を用いた省資源かつ効率的な感情推定手法の提案"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"感情認識・生体情報","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2024-11-11","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"法政大学理工学部"},{"subitem_text_value":"名古屋工業大学大学院工学研究科"},{"subitem_text_value":"芝浦工業大学電気電子情報専攻"},{"subitem_text_value":"芝浦工業大学電気電子情報専攻"}]},"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/240561/files/IPSJ-HCI24210020.pdf","label":"IPSJ-HCI24210020.pdf"},"date":[{"dateType":"Available","dateValue":"2026-11-11"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-HCI24210020.pdf","filesize":[{"value":"692.1 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":"33"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"755a9fa4-f468-4739-8bd9-a59a5d632983","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2024 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"福原, 陸翔"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"田中, 剛平"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"鈴木, 圭"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"菅谷, みどり"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Rikuto, Fukuhara","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Gouhei, Tanaka","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Kei, Suzuki","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Midori, Sugaya","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA1221543X","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":"2188-8760","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"近年，映画推薦やチャットボット，セラピーなど様々なサービスでユーザーの感情状態を推定し，活用することが期待されている．これらのサービスには，高精度な感情推定技術が貢献する．そのため，高い性能を示す深層学習を用いた感情推定研究が増加している．特に，無意識の生理反応を連続的かつ客観的な感情を定量化した値として取得できる方法として，心電・心拍データなどの生体情報を利用した手法が注目されている．しかし，これらの手法は潤沢な計算資源の使用を前提としており，高い計算コストを要する．そのため，スマートウォッチなどの計算資源が限られた身近なデバイスへの実装が困難であり，課題となっている．本研究では，本課題に対して，高精度かつ低計算コストで，省資源な感情推定手法の実現を目指す．具体的には，リザバーコンピューティング (Reservoir Computing) の一つである Echo State Network  (ESN) モデルを用い，深層学習と同等の精度を達成しつつ，より効率的な計算を実現する感情推定手法を提案する．この提案手法の有効性を検証するために，本研究では，DREAMERデータセットの心電 (ECG) データを使用し，提案手法である ESN モデルと深層学習に基づく Long Short-Term Memory (LSTM) モデルの性能を比較した．その結果，提案手法は LSTM モデルに比べ，精度は同等程度でありながら，学習時のメモリ使用量を約 52 分の 1，学習時間を約 25 分の 1 に削減した．本結果は提案手法の有効性，効率性を示しており，計算資源の限られた環境での感情推定に有望であるといえる．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"6","bibliographic_titles":[{"bibliographic_title":"研究報告ヒューマンコンピュータインタラクション（HCI）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2024-11-11","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"20","bibliographicVolumeNumber":"2024-HCI-210"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"links":{},"id":240561,"updated":"2025-01-19T07:56:25.503112+00:00"}