{"id":226929,"updated":"2025-01-19T12:20:10.355317+00:00","links":{},"created":"2025-01-19T01:26:14.989312+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00226929","sets":["934:10195:11103:11301"]},"path":["11301"],"owner":"44499","recid":"226929","title":["発話者の潜在ニーズ予測とその可視化Word2Vecモデルを用いた機械学習モデルの精度改善に関する検討"],"pubdate":{"attribute_name":"公開日","attribute_value":"2023-07-15"},"_buckets":{"deposit":"99dc3b06-479b-4707-8b6f-e01de979d1ad"},"_deposit":{"id":"226929","pid":{"type":"depid","value":"226929","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"発話者の潜在ニーズ予測とその可視化Word2Vecモデルを用いた機械学習モデルの精度改善に関する検討","author_link":["603522","603533","603530","603531","603527","603524","603529","603532","603526","603523","603528","603525"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"発話者の潜在ニーズ予測とその可視化Word2Vecモデルを用いた機械学習モデルの精度改善に関する検討"},{"subitem_title":"Prediction and Visualization of Latent Needs: Improving the Accuracy of Machine Learning Models using the Word2Vec Model","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"[一般投稿論文] 潜在ニーズ, 機械学習, 可視化, ニューラルネットワーク, Word2Vec","subitem_subject_scheme":"Other"}]},"item_type_id":"3","publish_date":"2023-07-15","item_3_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"医薬基盤・健康・栄養研究所"},{"subitem_text_value":"独立研究者"},{"subitem_text_value":"千葉大学医学部附属病院"},{"subitem_text_value":"医薬基盤・健康・栄養研究所"},{"subitem_text_value":"医薬品医療機器総合機構"},{"subitem_text_value":"医薬基盤・健康・栄養研究所"}]},"item_3_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"National Institutes of Biomedical Innovation, Health and Nutrition","subitem_text_language":"en"},{"subitem_text_value":"Independent Researcher","subitem_text_language":"en"},{"subitem_text_value":"Chiba University Hospital","subitem_text_language":"en"},{"subitem_text_value":"National Institutes of Biomedical Innovation, Health and Nutrition","subitem_text_language":"en"},{"subitem_text_value":"Pharmaceuticals and Medical Devices Agency","subitem_text_language":"en"},{"subitem_text_value":"National Institutes of Biomedical Innovation, Health and Nutrition","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/226929/files/IPSJ-TDP0403010.pdf","label":"IPSJ-TDP0403010.pdf"},"date":[{"dateType":"Available","dateValue":"2023-07-15"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-TDP0403010.pdf","filesize":[{"value":"1.2 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"5"},{"tax":["include_tax"],"price":"0","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"58fd0d45-a274-45bb-b3f7-e6032a8429bc","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2023 by the Information Processing Society of Japan"}]},"item_3_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"種村, 菜奈枝"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"町井, 湧介"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"佐々木, 剛"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"荒木, 通啓"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"佐藤, 淳子"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"千葉, 剛"}],"nameIdentifiers":[{}]}]},"item_3_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Nanae, Tanemura","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Yusuke, Machii","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Tsuyoshi, Sasaki","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Michihiro, Araki","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Junko, Sato","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Tsuyoshi, Chiba","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_3_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AA12894091","subitem_source_identifier_type":"NCID"}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_6501","resourcetype":"journal article"}]},"item_3_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2435-6484","subitem_source_identifier_type":"ISSN"}]},"item_3_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"近年，市民参画型の必要性は健康政策のみならず，社会における場でも議論されている．しかし，日本はハイコンテクスト文化であり，一般市民がコンテクストに頼らずに意見を明確な言葉として表現するコミュニケーションには不慣れであり，一般市民の声を政策等へ反映することは容易ではない．本研究では，口語テキストから発話者の潜在的ニーズを予測するための機械学習モデル構築，およびニューラルネットワークを用いて単語をベクトル変換する手法であるWord2Vecモデルを用いて機械学習モデルの精度改善を検討した．予備検討では，機械学習モデルの精度比較を行い，最適なモデルを選択した．本調査では，Word2Vecモデルを用いて同義語辞書を作成し，この辞書を使用して同一の特徴量に変換し学習を行う新手法を検討した．新手法の適応の有無で機械学習モデルの精度比較を行った．予備検討でのモデル選定実験の結果，モデル精度はxgboostでF値0.54と最も高く，本調査では，モデル精度は同義語辞書ありでF値0.61，なしでF値0.54であり，Word2Vecモデルを用いた同義語辞書の適応が機械学習モデルの精度改善に寄与した．","subitem_description_type":"Other"}]},"item_3_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"The need for public engagement has been deliberated in recent years. However, because of Japan's high context culture, Japanese people are not accustomed to communicating their opinions in clear. Therefore, it is not easy to reflect the voices of the general public in policy making. In the process of building a machine learning model for predicting the latent needs from spoken text, this study examined how to improve the accuracy of the model using the Word2Vec model, that uses a neural network to transform words into vectors. In this preliminary study, we compared the accuracy of machine learning models and selected the best model. We examined a new method that uses the Word2Vec model to create a synonym dictionary to convert the word clusters for identical features for learning. We compared the accuracy of machine learning models with and without adaptation of the dictionary. The results of model selection showed that xgboost had the highest model accuracy with an F value of 0.54. The model accuracy was 0.61 with the dictionary and 0.54 without. It showed that the adaptation of the synonym dictionary using the Word2Vec model can improve the accuracy of the model.","subitem_description_type":"Other"}]},"item_3_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"73","bibliographic_titles":[{"bibliographic_title":"情報処理学会論文誌デジタルプラクティス（DP）"}],"bibliographicPageStart":"69","bibliographicIssueDates":{"bibliographicIssueDate":"2023-07-15","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"3","bibliographicVolumeNumber":"4"}]},"relation_version_is_last":true,"weko_creator_id":"44499"}}