Music Emotion Classification of Chinese Songs based on Lyrics Using TF*IDF and Rhyme

This paper presents the outcomes of research into an automatic classification system based on the lingual part of music. Two novel kinds of short features are extracted from lyrics using tf*idf and rhyme. Meta-learning algorithm is adapted to combine these two sets of features. Results show that our features promote the accuracy of classification and meta-learning algorithm is effective in fusing the two features.

[1]  Ian H. Witten,et al.  The WEKA data mining software: an update , 2009, SKDD.

[2]  Andreas F. Ehmann,et al.  Lyric Text Mining in Music Mood Classification , 2009, ISMIR.

[3]  Yajie Hu,et al.  Lyric-based Song Emotion Detection with Affective Lexicon and Fuzzy Clustering Method , 2009, ISMIR.

[4]  J. Stephen Downie,et al.  When Lyrics Outperform Audio for Music Mood Classification: A Feature Analysis , 2010, ISMIR.

[5]  P. Laukka,et al.  Communication of emotions in vocal expression and music performance: different channels, same code? , 2003, Psychological bulletin.

[6]  Menno van Zaanen,et al.  Automatic Mood Classification Using TF*IDF Based on Lyrics , 2010, ISMIR.

[7]  Kam-Fai Wong,et al.  Sentiment vector space model for lyric-based song sentiment classification , 2008, ACL 2008.

[8]  Jun Wang,et al.  Predicting High-level Music Semantics Using Social Tags via Ontology-based Reasoning , 2010, ISMIR.

[9]  A. Bate,et al.  Poster session , 2009, 2009 Device Research Conference.

[10]  R. Thayer.  The biopsychology of mood and arousal , 1989 .

[11]  Jeffrey J. Scott,et al.  MUSIC EMOTION RECOGNITION: A STATE OF THE ART REVIEW , 2010 .

[12]  J. Russell.  A circumplex model of affect. , 1980 .

[13]  Jun Wang,et al.  Boosting for Multi-Modal Music Emotion Classification , 2010, ISMIR.

[14]  Venu Govindaraju,et al.  Review of Classifier Combination Methods , 2008, Machine Learning in Document Analysis and Recognition.

[15]  Ricardo Vilalta,et al.  A Perspective View and Survey of Meta-Learning , 2002, Artificial Intelligence Review.