{"id":207215,"updated":"2025-01-19T19:13:12.786748+00:00","links":{},"created":"2025-01-19T01:08:58.025409+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00207215","sets":["1164:8228:10133:10353"]},"path":["10353"],"owner":"44499","recid":"207215","title":["LSTMによる加速度に基づく個人推定"],"pubdate":{"attribute_name":"公開日","attribute_value":"2020-09-22"},"_buckets":{"deposit":"b15a65dd-897c-4f48-bb17-96f0890a9a08"},"_deposit":{"id":"207215","pid":{"type":"depid","value":"207215","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"LSTMによる加速度に基づく個人推定","author_link":["516739","516737","516741","516738","516740"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"LSTMによる加速度に基づく個人推定"}]},"item_type_id":"4","publish_date":"2020-09-22","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"工学院大学"},{"subitem_text_value":"工学院大学"},{"subitem_text_value":"長崎大学"},{"subitem_text_value":"お茶の水女子大学"},{"subitem_text_value":"工学院大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Kogakuin, University","subitem_text_language":"en"},{"subitem_text_value":"Kogakuin, University","subitem_text_language":"en"},{"subitem_text_value":"Nagasaki University","subitem_text_language":"en"},{"subitem_text_value":"Ochanomizu University","subitem_text_language":"en"},{"subitem_text_value":"Kogakuin, 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/207215/files/IPSJ-ASD20019002.pdf","label":"IPSJ-ASD20019002.pdf"},"date":[{"dateType":"Available","dateValue":"2022-09-22"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-ASD20019002.pdf","filesize":[{"value":"738.6 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":"49"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"ef52ef73-0aa4-4d20-b2db-9b11b804f870","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2020 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"高橋, 良颯"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"中村, 鴻介"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"神山, 剛"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"小口, 正人"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"山口, 実靖"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA1271737X","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":"2189-4450","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"スマートフォンの加速度センサからえられた加速度データを深層学習 (Deep Neural Network) を用いて解析し，スマートフォンの保持者を推定する手法が提案されている．ただし，当該手法は各時刻に得られた加速度データをソートし DNN に入力する方法をとっており，時系列データである加速度データ列の時間的特徴を考慮できない課題があり，正答率も 8 割強にとどまっている．本稿では，加速度データを LSTM (Long short-term memory) を用いて解析し，保持者を推定する手法を 2 つ提案する．一つは，事前データを LSTM により学習し推定対象データを分類する手法であり，最も確率が高いと出力されたユーザを推定結果とする．もう一つは，出力確率が閾値以下である場合は推定結果を不明として false positive の発生確率を抑える手法である．そして，実ユーザ 5 人による評価実験の結果を示し，高い精度でユーザを推定できることを示す．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"7","bibliographic_titles":[{"bibliographic_title":"研究報告高齢社会デザイン（ASD）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2020-09-22","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"2","bibliographicVolumeNumber":"2020-ASD-19"}]},"relation_version_is_last":true,"weko_creator_id":"44499"}}