Using machine learning analysis to interpret the relationship between music emotion and lyric features

Melody and lyrics, reflecting two unique human cognitive abilities, are usually combined in music to convey emotions. Although psychologists and computer scientists have made considerable progress in revealing the association between musical structure and the perceived emotions of music, the feature...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Liang Xu, Zaoyi Sun, Xin Wen, Zhengxi Huang, Chi-ju Chao, Liuchang Xu
Formato:	article
Lenguaje:	EN
Publicado:	PeerJ Inc. 2021
Materias:	Music emotion recognition Lyric feature extraction Audio signal processing LIWC Chinese pop song Electronic computers. Computer science QA75.5-76.95
Acceso en línea:	https://doaj.org/article/d06bf0c177db4209aaaba3f632736290
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:d06bf0c177db4209aaaba3f632736290
record_format	dspace
spelling	oai:doaj.org-article:d06bf0c177db4209aaaba3f6327362902021-11-17T15:05:18ZUsing machine learning analysis to interpret the relationship between music emotion and lyric features10.7717/peerj-cs.7852376-5992https://doaj.org/article/d06bf0c177db4209aaaba3f6327362902021-11-01T00:00:00Zhttps://peerj.com/articles/cs-785.pdfhttps://peerj.com/articles/cs-785/https://doaj.org/toc/2376-5992Melody and lyrics, reflecting two unique human cognitive abilities, are usually combined in music to convey emotions. Although psychologists and computer scientists have made considerable progress in revealing the association between musical structure and the perceived emotions of music, the features of lyrics are relatively less discussed. Using linguistic inquiry and word count (LIWC) technology to extract lyric features in 2,372 Chinese songs, this study investigated the effects of LIWC-based lyric features on the perceived arousal and valence of music. First, correlation analysis shows that, for example, the perceived arousal of music was positively correlated with the total number of lyric words and the mean number of words per sentence and was negatively correlated with the proportion of words related to the past and insight. The perceived valence of music was negatively correlated with the proportion of negative emotion words. Second, we used audio and lyric features as inputs to construct music emotion recognition (MER) models. The performance of random forest regressions reveals that, for the recognition models of perceived valence, adding lyric features can significantly improve the prediction effect of the model using audio features only; for the recognition models of perceived arousal, lyric features are almost useless. Finally, by calculating the feature importance to interpret the MER models, we observed that the audio features played a decisive role in the recognition models of both perceived arousal and perceived valence. Unlike the uselessness of the lyric features in the arousal recognition model, several lyric features, such as the usage frequency of words related to sadness, positive emotions, and tentativeness, played important roles in the valence recognition model.Liang XuZaoyi SunXin WenZhengxi HuangChi-ju ChaoLiuchang XuPeerJ Inc.articleMusic emotion recognitionLyric feature extractionAudio signal processingLIWCChinese pop songElectronic computers. Computer scienceQA75.5-76.95ENPeerJ Computer Science, Vol 7, p e785 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Music emotion recognition Lyric feature extraction Audio signal processing LIWC Chinese pop song Electronic computers. Computer science QA75.5-76.95
spellingShingle	Music emotion recognition Lyric feature extraction Audio signal processing LIWC Chinese pop song Electronic computers. Computer science QA75.5-76.95 Liang Xu Zaoyi Sun Xin Wen Zhengxi Huang Chi-ju Chao Liuchang Xu Using machine learning analysis to interpret the relationship between music emotion and lyric features
description	Melody and lyrics, reflecting two unique human cognitive abilities, are usually combined in music to convey emotions. Although psychologists and computer scientists have made considerable progress in revealing the association between musical structure and the perceived emotions of music, the features of lyrics are relatively less discussed. Using linguistic inquiry and word count (LIWC) technology to extract lyric features in 2,372 Chinese songs, this study investigated the effects of LIWC-based lyric features on the perceived arousal and valence of music. First, correlation analysis shows that, for example, the perceived arousal of music was positively correlated with the total number of lyric words and the mean number of words per sentence and was negatively correlated with the proportion of words related to the past and insight. The perceived valence of music was negatively correlated with the proportion of negative emotion words. Second, we used audio and lyric features as inputs to construct music emotion recognition (MER) models. The performance of random forest regressions reveals that, for the recognition models of perceived valence, adding lyric features can significantly improve the prediction effect of the model using audio features only; for the recognition models of perceived arousal, lyric features are almost useless. Finally, by calculating the feature importance to interpret the MER models, we observed that the audio features played a decisive role in the recognition models of both perceived arousal and perceived valence. Unlike the uselessness of the lyric features in the arousal recognition model, several lyric features, such as the usage frequency of words related to sadness, positive emotions, and tentativeness, played important roles in the valence recognition model.
format	article
author	Liang Xu Zaoyi Sun Xin Wen Zhengxi Huang Chi-ju Chao Liuchang Xu
author_facet	Liang Xu Zaoyi Sun Xin Wen Zhengxi Huang Chi-ju Chao Liuchang Xu
author_sort	Liang Xu
title	Using machine learning analysis to interpret the relationship between music emotion and lyric features
title_short	Using machine learning analysis to interpret the relationship between music emotion and lyric features
title_full	Using machine learning analysis to interpret the relationship between music emotion and lyric features
title_fullStr	Using machine learning analysis to interpret the relationship between music emotion and lyric features
title_full_unstemmed	Using machine learning analysis to interpret the relationship between music emotion and lyric features
title_sort	using machine learning analysis to interpret the relationship between music emotion and lyric features
publisher	PeerJ Inc.
publishDate	2021
url	https://doaj.org/article/d06bf0c177db4209aaaba3f632736290
work_keys_str_mv	AT liangxu usingmachinelearninganalysistointerprettherelationshipbetweenmusicemotionandlyricfeatures AT zaoyisun usingmachinelearninganalysistointerprettherelationshipbetweenmusicemotionandlyricfeatures AT xinwen usingmachinelearninganalysistointerprettherelationshipbetweenmusicemotionandlyricfeatures AT zhengxihuang usingmachinelearninganalysistointerprettherelationshipbetweenmusicemotionandlyricfeatures AT chijuchao usingmachinelearninganalysistointerprettherelationshipbetweenmusicemotionandlyricfeatures AT liuchangxu usingmachinelearninganalysistointerprettherelationshipbetweenmusicemotionandlyricfeatures
_version_	1718425447016955904

Using machine learning analysis to interpret the relationship between music emotion and lyric features

Ejemplares similares