TY - GEN
T1 - Detecting emotional expression of music with feature selection approach
AU - Hwang, Fang Chen
AU - Wang, Jeen-Shing
AU - Chung, Pau-Choo
AU - Yang, Ching Fang
PY - 2013/7/12
Y1 - 2013/7/12
N2 - This paper presents a mechanism on detecting emotional expression of music with feature selection approach. Happiness, sadness, anger, and peace are considered in the classification problem. The thirty-seven features were extracted to represent the characteristics of music samples, such as rhythm, dynamic, pitch, and timbre features. The kernel-based class separability (KBCS) was introduced to prioritize features for emotion classification because not all features have the same importance in achieving emotional expression. Two feature transformation techniques, principal component analysis (PCA) and linear discriminant analysis (LDA) were applied after the feature selection. The inclusion of these two techniques can effectively improve the classification accuracy. To the end, the k-nearest neighborhood (k-NN) classifier is adopted. The results indicate that the proposed method in the study can achieve accuracy at almost 90%.
AB - This paper presents a mechanism on detecting emotional expression of music with feature selection approach. Happiness, sadness, anger, and peace are considered in the classification problem. The thirty-seven features were extracted to represent the characteristics of music samples, such as rhythm, dynamic, pitch, and timbre features. The kernel-based class separability (KBCS) was introduced to prioritize features for emotion classification because not all features have the same importance in achieving emotional expression. Two feature transformation techniques, principal component analysis (PCA) and linear discriminant analysis (LDA) were applied after the feature selection. The inclusion of these two techniques can effectively improve the classification accuracy. To the end, the k-nearest neighborhood (k-NN) classifier is adopted. The results indicate that the proposed method in the study can achieve accuracy at almost 90%.
UR - http://www.scopus.com/inward/record.url?scp=84879874398&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84879874398&partnerID=8YFLogxK
U2 - 10.1109/ICOT.2013.6521213
DO - 10.1109/ICOT.2013.6521213
M3 - Conference contribution
AN - SCOPUS:84879874398
SN - 9781467359368
T3 - ICOT 2013 - 1st International Conference on Orange Technologies
SP - 282
EP - 286
BT - ICOT 2013 - 1st International Conference on Orange Technologies
T2 - 1st International Conference on Orange Technologies, ICOT 2013
Y2 - 12 March 2013 through 16 March 2013
ER -