File Download
  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: A study on cross-cultural and cross-dataset generalizability of music mood regression models

TitleA study on cross-cultural and cross-dataset generalizability of music mood regression models
Authors
Issue Date2014
Citation
The 40th International Computer Music Conference (ICMC 2014) Joint with the 11th Sound and Music Computing Conference (SMC 2014), Athens, Greece, 14-20 September 2014. In Conference Proceedings, 2014, p. 1149-1155 How to Cite?
AbstractThe goal of music mood regression is to represent the emotional expression of music pieces as numerical values in a low-dimensional mood space and automatically predict those values for unseen music pieces. Existing studies on this topic usually train and test regression models using music datasets sampled from the same culture source, annotated by people with the same cultural background, or otherwise constructed by the same method. In this study, we explore whether and to what extent regression models trained with samples in one dataset can be applied to predicting valence and arousal values of samples in another dataset. Specifically, three datasets that differ in factors such as cultural backgrounds of stimuli (music) and subjects (annotators), stimulus types and annotation methods are evaluated and the results suggested that cross-cultural and cross-dataset predictions of both valence and arousal values could achieve comparable performance to within-dataset predictions. We also discuss how the generalizability of regression models can be affected by dataset characteristics. Findings of this study may provide valuable insights into music mood regression for non-Western and other music where training data are scarce. © 2014 Xiao Hu et al.
DescriptionConference theme: Music Technology Meets Philosophy: From Digital Echos to Virtual Ethos
Persistent Identifierhttp://hdl.handle.net/10722/213520
ISBN

 

DC FieldValueLanguage
dc.contributor.authorHu, X-
dc.contributor.authorYang, YH-
dc.date.accessioned2015-08-04T04:47:50Z-
dc.date.available2015-08-04T04:47:50Z-
dc.date.issued2014-
dc.identifier.citationThe 40th International Computer Music Conference (ICMC 2014) Joint with the 11th Sound and Music Computing Conference (SMC 2014), Athens, Greece, 14-20 September 2014. In Conference Proceedings, 2014, p. 1149-1155-
dc.identifier.isbn978-960466137-4-
dc.identifier.urihttp://hdl.handle.net/10722/213520-
dc.descriptionConference theme: Music Technology Meets Philosophy: From Digital Echos to Virtual Ethos-
dc.description.abstractThe goal of music mood regression is to represent the emotional expression of music pieces as numerical values in a low-dimensional mood space and automatically predict those values for unseen music pieces. Existing studies on this topic usually train and test regression models using music datasets sampled from the same culture source, annotated by people with the same cultural background, or otherwise constructed by the same method. In this study, we explore whether and to what extent regression models trained with samples in one dataset can be applied to predicting valence and arousal values of samples in another dataset. Specifically, three datasets that differ in factors such as cultural backgrounds of stimuli (music) and subjects (annotators), stimulus types and annotation methods are evaluated and the results suggested that cross-cultural and cross-dataset predictions of both valence and arousal values could achieve comparable performance to within-dataset predictions. We also discuss how the generalizability of regression models can be affected by dataset characteristics. Findings of this study may provide valuable insights into music mood regression for non-Western and other music where training data are scarce. © 2014 Xiao Hu et al.-
dc.languageeng-
dc.relation.ispartofProceedings ICMC / SMC 2014-
dc.rightsCreative Commons: Attribution 3.0 Hong Kong License-
dc.titleA study on cross-cultural and cross-dataset generalizability of music mood regression models-
dc.typeConference_Paper-
dc.identifier.emailHu, X: xiaoxhu@hku.hk-
dc.identifier.authorityHu, X=rp01711-
dc.description.naturepublished_or_final_version-
dc.identifier.scopuseid_2-s2.0-84908874328-
dc.identifier.hkuros246064-
dc.identifier.spage1149-
dc.identifier.epage1155-
dc.customcontrol.immutablesml 150804-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats