Rough Set-Based Approach for Automatic Emotion Classification of Music = Rough Set-Based Approach for Automatic Emotion Classification of Music

다국어 초록 (Multilingual Abstract)

Music emotion is an important component in the field of music information retrieval and computationalmusicology. This paper proposes an approach for automatic emotion classification, based on rough set (RS)theory. In the proposed approach, four different sets of music features are extracted, representing dynamics,rhythm, spectral, and harmony. From the features, five different statistical parameters are considered asattributes, including up to the 4^th order central moments of each feature, and covariance components ofmutual ones. The large number of attributes is controlled by RS-based approach, in which superfluousfeatures are removed, to obtain indispensable ones. In addition, RS-based approach makes it possible tovisualize which attributes play a significant role in the generated rules, and also determine the strength of eachrule for classification. The experiments have been performed to find out which audio features and which ofthe different statistical parameters derived from them are important for emotion classification. Also, theresulting indispensable attributes and the usefulness of covariance components have been discussed. Theoverall classification accuracy with all statistical parameters has recorded comparatively better than currentlyexisting methods on a pair of datasets.

참고문헌 (Reference)

1 "Soundtracks datasets for music and emotion"

2 S. Dalai, "Rough-set-based feature selection and classification for power quality sensing device employing correlation techniques" 13 (13): 563-573, 2013

3 Z. Pawlak, "Rough set theory and its applications" 9 (9): 7-10, 2002

4 Z. Pawlak, "Rough Sets: Theoretical Aspects of Reasoning about Data" Kluwer Academic Publisher 1991

5 A. Sen, "Regression Analysis: Theory, Methods, and Applications" Springer 1990

6 Y. H. Yang, "Ranking-based emotion recognition for music organization and retrieval" 19 (19): 762-774, 2011

7 "ROSE2 (Rough Sets Data Explorer)"

8 D. Cabrera, "PsySound: a computer program for psychoacoustical analysis" 47-54, 1999

9 A. Papoulis, "Pillai, Probability, Random Variables, and Stochastic Processes" McGraw-Hill 2002

10 G. Tzanetakis, "Musical genre classification of audio signals" 10 (10): 293-302, 2002

11 B. K. Baniya, "Music mood classification using reduced audio features" 915-917, 2015

12 R. Panda, "Multi-modal music emotion recognition: a new dataset, methodology and comparative analysis" 1-13, 2013

13 "MOODetector"

14 O. Lartillot, "MIR in MATLAB (II): a toolbox for musical feature extraction from audio" 127-130, 2007

15 F. Pachet, "Improving multilabel analysis of music titles : a large-scale validation of the correction approach" 17 (17): 335-343, 2009

16 P. Saari, "Generalizability and simplicity as criteria in feature selection : application to mood classification in music" 19 (19): 1802-1812, 2011

17 B. K. Baniya, "Evaluation of different audio features for musical genre classification" 260-265, 2013

18 H. S. Nguyen, "Discretization of real value attributes: a Boolean reasoning approach" Warsaw University 1997

19 T. Li, "Content-based music similarity search and emotion detection" 705-708, 2004

20 B. K. Baniya, "Automatic music genre classification using timbral texture and rhythmic content features" 434-443, 2015

21 L. Lu, "Automatic mood detection and tracking of music audio signals" 14 (14): 5-18, 2006

22 B. K. Baniya, "Audio feature reduction and analysis for automatic music genre classification" Man and Cybernetics (SMC) 457-462, 2004

23 J. A. Russell, "Affect grid : a single-item scale of pleasure and arousal" 57 (57): 493-502, 1989

24 D. P. Solomatine, "AdaBoost. RT : a boosting algorithm for regression problems" 1163-1168, 2004

25 A. J. Smola, "A tutorial on support vector regression" 14 (14): 199-222, 2004

26 Y. H. Yang, "A regression approach to music emotion recognition" 16 (16): 448-457, 2008

27 E. Pampalk, "A MATLAB toolbox to compute music similarity from audio" 1-4, 2004