Tuyển tập báo cáo các nghiên cứu khoa học quốc tế ngành hóa học dành cho các bạn yêu hóa học tham khảo đề tài: Multi-label classification of music by emotion | Trohidis et al. EURASIP Journal on Audio Speech and Music Processing 2011 2011 4 http content 2011 1 4 D EURASIP Journal on Audio Speech and Music Processing a SpringerOpen Journal RESEARCH Open Access Multi-label classification of music by emotion 1 2 1 2 Konstantinos Trohidis Grigorios Tsoumakas George Kalliris and loannis Vlahavas Abstract This work studies the task of automatic emotion detection in music. Music may evoke more than one different emotion at the same time. Single-label classification and regression cannot model this multiplicity. Therefore this work focuses on multi-label classification approaches where a piece of music may simultaneously belong to more than one class. Seven algorithms are experimentally compared for this task. Furthermore the predictive power of several audio features is evaluated using a new multi-label feature selection method. Experiments are conducted on a set of 593 songs with six clusters of emotions based on the Tellegen-Watson-Clark model of affect. Results show that multi-label modeling is successful and provide interesting insights into the predictive quality of the algorithms and features. Keywords multi-label classification feature selection music information retrieval 1. Introduction Humans by nature are emotionally affected by music. Who can argue against the famous quote of the German philosopher Friedrich Nietzsche who said that without music life would be a mistake . 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 pro-gramsa 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 6-9 task. Music may evoke more than one different emotion at the same time. Single-label classification and
