In this paper we propose a feature set for emotion classification of Western popular music. We show that by surveying a range of common feature extraction methods, a set of five features can model emotion with good accuracy. To evaluate the system we implement an independent feature evaluation paradigm aimed at testing the property of generalizability; the ability of a machine learning algorithm to maintain good performance over different data sets.
|Title of host publication||The 9th International Symposium on Computer Music Modeling and Retrieval (CMMR) Music and Emotions|
|Number of pages||10|
|Publication status||Published - Jun 2012|
|Name||Lecture Notes in Computer Sciences |
- music emotion classification
- popular music
- support vector machine