WEKO3
アイテム
{"_buckets": {"deposit": "b81844c6-8c94-4e8d-b13a-64f37a253dcf"}, "_deposit": {"created_by": 4, "id": "4339", "owners": [4], "pid": {"revision_id": 0, "type": "depid", "value": "4339"}, "status": "published"}, "_oai": {"id": "oai:naist.repo.nii.ac.jp:00004339", "sets": ["35"]}, "author_link": ["10110", "10111", "10112"], "item_7_biblio_info_9": {"attribute_name": "書誌情報", "attribute_value_mlt": [{"bibliographicIssueDates": {"bibliographicIssueDate": "2021-08-03", "bibliographicIssueDateType": "Issued"}, "bibliographicVolumeNumber": "7", "bibliographic_titles": [{"bibliographic_title": "PeerJ Computer Science", "bibliographic_titleLang": "en"}]}]}, "item_7_description_7": {"attribute_name": "抄録", "attribute_value_mlt": [{"subitem_description": "Background: Preferences for music can be represented through music features. Thewidespread prevalence of music streaming has allowed for music feature informationto be consolidated by service providers like Spotify. In this paper, we demonstratethat machine learning classification on cultural market membership (Taiwanese,Japanese, American) by music features reveals variations in popular music acrossthese markets.Methods: We present an exploratory analysis of 1.08 million songs centred onTaiwanese, Japanese and American markets. We use both multiclass classificationmodels (Gradient Boosted Decision Trees (GBDT) and Multilayer Perceptron(MLP)), and binary classification models, and interpret their results using variableimportance measures and Partial Dependence Plots. To ensure the reliability of ourinterpretations, we conducted a follow-up study comparing Top-50 playlists fromTaiwan, Japan, and the US on identified variables of importance.Results: The multiclass models achieved moderate classification accuracy (GBDT =0.69, MLP = 0.66). Accuracy scores for binary classification models ranged between0.71 to 0.81. Model interpretation revealed music features of greatest importance:Overall, popular music in Taiwan was characterised by high acousticness, Americanmusic was characterised by high speechiness, and Japanese music was characterisedby high energy features. A follow-up study using Top-50 charts found similarlysignificant differences between cultures for these three features.Conclusion: We demonstrate that machine learning can reveal both the magnitudeof differences in music preference across Taiwanese, Japanese, and Americanmarkets, and where these preferences are different. While this paper is limited toSpotify data, it underscores the potential contribution of machine learning inexploratory approaches to research on cultural differences.", "subitem_description_language": "en", "subitem_description_type": "Abstract"}]}, "item_7_publisher_10": {"attribute_name": "出版者", "attribute_value_mlt": [{"subitem_publisher": "PeerJ", "subitem_publisher_language": "en"}]}, "item_7_relation_17": {"attribute_name": "出版者版DOI", "attribute_value_mlt": [{"subitem_relation_type": "isReplacedBy", "subitem_relation_type_id": {"subitem_relation_type_id_text": "https://doi.org/10.7717/peerj-cs.642", "subitem_relation_type_select": "DOI"}}]}, "item_7_relation_22": {"attribute_name": "出版者版URI", "attribute_value_mlt": [{"subitem_relation_type": "isReplacedBy", "subitem_relation_type_id": {"subitem_relation_type_id_text": "https://peerj.com/articles/cs-642/", "subitem_relation_type_select": "URI"}}]}, "item_7_rights_18": {"attribute_name": "権利", "attribute_value_mlt": [{"subitem_rights": "c 2021 Liew et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Computer Science) and either DOI or URL of the article must be cited.", "subitem_rights_language": "en"}]}, "item_7_source_id_12": {"attribute_name": "EISSN/PISSN", "attribute_value_mlt": [{"subitem_source_identifier": "2376-5992", "subitem_source_identifier_type": "ISSN"}]}, "item_7_text_25": {"attribute_name": "NAIST ID", "attribute_value_mlt": [{"subitem_text_value": "74656828"}]}, "item_access_right": {"attribute_name": "アクセス権", "attribute_value_mlt": [{"subitem_access_right": "metadata only access", "subitem_access_right_uri": "http://purl.org/coar/access_right/c_14cb"}]}, "item_creator": {"attribute_name": "著者", "attribute_type": "creator", "attribute_value_mlt": [{"creatorNames": [{"creatorName": "Liew, Kongmeng", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "10110", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Uchida, Yukiko", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "10111", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "de, Almeida Igor", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "10112", "nameIdentifierScheme": "WEKO"}]}]}, "item_keyword": {"attribute_name": "キーワード", "attribute_value_mlt": [{"subitem_subject": "Music", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Culture", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Psychology", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Spotify", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Machine Learning", "subitem_subject_language": "en", "subitem_subject_scheme": "Other"}]}, "item_language": {"attribute_name": "言語", "attribute_value_mlt": [{"subitem_language": "eng"}]}, "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": "Cultural differences in music features across Taiwanese, Japanese and American markets", "item_titles": {"attribute_name": "タイトル", "attribute_value_mlt": [{"subitem_title": "Cultural differences in music features across Taiwanese, Japanese and American markets", "subitem_title_language": "en"}]}, "item_type_id": "7", "owner": "4", "path": ["35"], "permalink_uri": "http://hdl.handle.net/10061/14567", "pubdate": {"attribute_name": "PubDate", "attribute_value": "2021-12-02"}, "publish_date": "2021-12-02", "publish_status": "0", "recid": "4339", "relation": {}, "relation_version_is_last": true, "title": ["Cultural differences in music features across Taiwanese, Japanese and American markets"], "weko_shared_id": -1}
Cultural differences in music features across Taiwanese, Japanese and American markets
http://hdl.handle.net/10061/14567
http://hdl.handle.net/10061/14567508b36a9-6cba-4fa9-b9dc-9dba8a3730cd
Item type | 学術雑誌論文 / Journal Article(1) | |||||
---|---|---|---|---|---|---|
公開日 | 2021-12-02 | |||||
タイトル | ||||||
タイトル | Cultural differences in music features across Taiwanese, Japanese and American markets | |||||
言語 | ||||||
言語 | eng | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | Music | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | Culture | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | Psychology | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | Spotify | |||||
キーワード | ||||||
主題Scheme | Other | |||||
主題 | Machine Learning | |||||
資源タイプ | ||||||
資源タイプ | journal article | |||||
アクセス権 | ||||||
アクセス権 | metadata only access | |||||
著者 |
Liew, Kongmeng
× Liew, Kongmeng× Uchida, Yukiko× de, Almeida Igor |
|||||
抄録 | ||||||
内容記述タイプ | Abstract | |||||
内容記述 | Background: Preferences for music can be represented through music features. Thewidespread prevalence of music streaming has allowed for music feature informationto be consolidated by service providers like Spotify. In this paper, we demonstratethat machine learning classification on cultural market membership (Taiwanese,Japanese, American) by music features reveals variations in popular music acrossthese markets.Methods: We present an exploratory analysis of 1.08 million songs centred onTaiwanese, Japanese and American markets. We use both multiclass classificationmodels (Gradient Boosted Decision Trees (GBDT) and Multilayer Perceptron(MLP)), and binary classification models, and interpret their results using variableimportance measures and Partial Dependence Plots. To ensure the reliability of ourinterpretations, we conducted a follow-up study comparing Top-50 playlists fromTaiwan, Japan, and the US on identified variables of importance.Results: The multiclass models achieved moderate classification accuracy (GBDT =0.69, MLP = 0.66). Accuracy scores for binary classification models ranged between0.71 to 0.81. Model interpretation revealed music features of greatest importance:Overall, popular music in Taiwan was characterised by high acousticness, Americanmusic was characterised by high speechiness, and Japanese music was characterisedby high energy features. A follow-up study using Top-50 charts found similarlysignificant differences between cultures for these three features.Conclusion: We demonstrate that machine learning can reveal both the magnitudeof differences in music preference across Taiwanese, Japanese, and Americanmarkets, and where these preferences are different. While this paper is limited toSpotify data, it underscores the potential contribution of machine learning inexploratory approaches to research on cultural differences. | |||||
書誌情報 |
en : PeerJ Computer Science 巻 7, 発行日 2021-08-03 |
|||||
出版者 | ||||||
出版者 | PeerJ | |||||
EISSN/PISSN | ||||||
収録物識別子タイプ | ISSN | |||||
収録物識別子 | 2376-5992 | |||||
出版者版DOI | ||||||
関連タイプ | isReplacedBy | |||||
識別子タイプ | DOI | |||||
関連識別子 | https://doi.org/10.7717/peerj-cs.642 | |||||
出版者版URI | ||||||
関連タイプ | isReplacedBy | |||||
識別子タイプ | URI | |||||
関連識別子 | https://peerj.com/articles/cs-642/ | |||||
権利 | ||||||
権利情報 | c 2021 Liew et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Computer Science) and either DOI or URL of the article must be cited. |