Music Emotions Recognition by Machine Learning With Cognitive Classification Methodologies

Music Emotions Recognition by Machine Learning With Cognitive Classification Methodologies

Junjie Bai (School of Electrical and Information Engineering, Chongqing University of Science and Technology, Chongqing, China & Schulich School of Engineering and Hotchkiss Brain Institute, University of Calgary, Calgary, Canada), Kan Luo (School of Information Science and Engineering, Fujian University of Technology, Fuzhou, China), Jun Peng (School of Electrical and Information Engineering, Chongqing University of Science and Technology, Chongqing, China), Jinliang Shi (School of Electrical and Information Engineering, Chongqing University of Science and Technology, Chongqing, China), Ying Wu (School of Electrical and Information Engineering, Chongqing University of Science and Technology, Chongqing, China), Lixiao Feng (School of Electrical and Information Engineering, Chongqing University of Science and Technology, Chongqing, China & Schulich School of Engineering and Hotchkiss Brain Institute, University of Calgary, Calgary, Canada), Jianqing Li (School of Instrument Science and Engineering, Southeast University, Nanjing, China) and Yingxu Wang (International Institute of Cognitive Informatics and Cognitive Computing (ICIC), Laboratory for Computational Intelligence, Denotational Mathematics and Software Science, Department of Electrical and Computer Engineering, Schulich School of Engineering and Hotchkiss Brain Institute, University of Calgary, Calgary, Canada & Information Systems Lab, Stanford University, Stanford, USA)
DOI: 10.4018/978-1-7998-2460-2.ch052

Abstract

Music emotions recognition (MER) is a challenging field of studies addressed in multiple disciplines such as musicology, cognitive science, physiology, psychology, arts and affective computing. In this article, music emotions are classified into four types known as those of pleasing, angry, sad and relaxing. MER is formulated as a classification problem in cognitive computing where 548 dimensions of music features are extracted and modeled. A set of classifications and machine learning algorithms are explored and comparatively studied for MER, which includes Support Vector Machine (SVM), k-Nearest Neighbors (KNN), Neuro-Fuzzy Networks Classification (NFNC), Fuzzy KNN (FKNN), Bayes classifier and Linear Discriminant Analysis (LDA). Experimental results show that the SVM, FKNN and LDA algorithms are the most effective methodologies that obtain more than 80% accuracy for MER.
Chapter Preview
Top

2.1. The Emotion Plane and the Valence-Arousal Emotion Model

Russell and Thayer proposed the valence-arousal model for music emotion description in 1980s (Russell, 1980) which is widely used by musicologists, estheticians and psychologists. A 2-D emotion plane was introduced in the dimensions of valence and arousal (VA). In the VA plane, the horizontal axis represents the valence value in positive or negative emotion. The vertical axis denotes the arousal value of exciting or calming. Both VA values are defined in the range [-1, 1]. In this measurement scope, a higher valence value means a higher and positive emotions, and a higher arousal value indicates a stronger emotional intensity.

Figure 1.

The valance-arousal plane of music emotions

978-1-7998-2460-2.ch052.f01

Figure 1 provides four quadrants in the music emotion plane corresponding to the categories of positive and strong (I), negative and strong (II), negative and weak (III), as well as positive and weak (IV) as described in Table 1.

Table 1.
The cognitive quadrants in the music emotion plane
QuadrantEmotional Type
IExcited, happy, and pleased
IIAnnoying, angry, and nervous
IIISad, bored, and sleepy
IVRelaxed, peaceful, and calm

According to Figure 1, a happy emotional experience is an emotion of positive valence and highly arousal, while sad is an emotion of negative valence and low arousal. Therefore, any form of music emotions can be mapped onto a certain point in the VA plane. This allows music emotions to be formally recognized by a pair of VA values (Thayer, 1989). The dimensional model of music emotions provides a simple, reliable and cognitive model for practical affective experiments and manipulations.

Complete Chapter List

Search this Book:
Reset