Advances in Soft Computing
DOI: 10.1007/3-540-33521-8_30
|View full text |Cite
|
Sign up to set email alerts
|

Abstract: Abstract. This paper addresses the problem of multi-label classification of emotions in musical recordings. The data set contains 875 samples (30 seconds each). The samples were manually labelled into 13 classes, without limits regarding the number of labels for each sample. The experiments and the results are discussed in this paper.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
24
0
1

Publication Types

Select...
5
3
1

Relationship

0
9

Authors

Journals

citations
Cited by 80 publications
(25 citation statements)
references
References 13 publications
0
24
0
1
Order By: Relevance
“…The same 13 clusters as in [6] were used in [8], where the authors modified the k Nearest Neighbors algorithm in order to handle multi-label data directly. They found that the predictive performance was low, too.…”
Section: Multi-label Classificationmentioning
confidence: 99%
See 1 more Smart Citation
“…The same 13 clusters as in [6] were used in [8], where the authors modified the k Nearest Neighbors algorithm in order to handle multi-label data directly. They found that the predictive performance was low, too.…”
Section: Multi-label Classificationmentioning
confidence: 99%
“…As music databases grow in size and number, the retrieval of music by emotion is becoming an important task for various applications, such as song selection in mobile devices [1], music recommendation systems [2], TV and radio programs a , and music therapy. Past approaches towards automated detection of emotions in music modeled the learning problem as a single-label classification [3,4] regression [5], or multi-label classification [6][7][8][9] task. Music may evoke more than one different emotion at the same time.…”
Section: Introductionmentioning
confidence: 99%
“…Apesar de grande parte das pesquisas na área de classificação de dados se concentrarem na análise de dados com um único rótulo, nos últimos anos, a área de classificação de dados multirrótulo tem atraído a atenção de grande parte da comunidade científica, motivado principalmente pelo grande aumento no número de aplicações, tais como para classificação de proteínas [21]; mineração de mídias sociais Estratégias de Construções de Comitês de Classificadores Multirrótulos no Aprendizado Semissupervisionado Multidescrição [22]; avaliação semântica da integração da gestão de riscos de segurança em documentos de software da administração pública [23]; função genômica [24,25]; categorização de músicas através de emoções [26][27][28][29] e marketing direcionado [30].…”
Section: Classificação De Dados Multirrótulounclassified
“…On comparing different audio feature sets, the highest accuracy achieved was found to be 50% by combining spectral contrast and MFCC features. Several studies have argued that one mood category is not always enough to represent the mood of a music piece, and, thus, music mood classification may better be formulated as a multi-label classification problem (Wieczorkowska et al, 2006;Trohidis et al, 2008).…”
Section: Exploiting Listener Ratingsmentioning
confidence: 99%