摘要: Music emotion recognition today is based on techniques that require high quality and large emotionally labeled sets of songs to train algorithms. Manual professional annotations are costly hardly accomplished. There a need for datasets public, highly polarized, in size following popular representation models. In this paper we present the steps followed create two such using intelligence last.fm community tags. first dataset, categorized an space four clusters adopted from literature observations. The second dataset discriminates between positive negative only. We also observed mood tags biased towards emotions. This imbalance was reflected cluster sizes resulting obtained; they contain more than ones.
core.ac.uk
sci-hub.st 参考文章(30)
Cyril Laurier, Perfecto Herrera, Mohamed Sordo, Joan Serrà, MUSIC MOOD REPRESENTATIONS FROM SOCIAL TAGS international symposium/conference on music information retrieval. pp. 381- 386 ,(2009)
Raluca Paiu, Claudiu S. Firan, Cyril Laurier, Kerstin Bischoff, Mohamed Sordo, Wolfgang Nejdl, MUSIC MOOD AND THEME CLASSIFICATION - A HYBRID APPROACH international symposium/conference on music information retrieval. pp. 657- 662 ,(2009)
Andreas F. Ehmann, J. Stephen Downie, Xiao Hu, LYRIC TEXT MINING IN MUSIC MOOD CLASSIFICATION international symposium/conference on music information retrieval. pp. 411- 416 ,(2009)
J. Stephen Downie, Mert Bay, Xiao Hu, CREATING A SIMPLIFIED MUSIC MOOD CLASSIFICATION GROUND-TRUTH SET international symposium/conference on music information retrieval. pp. 309- 310 ,(2007)
Hui He, Jianming Jin, Yuhong Xiong, Bo Chen, Wu Sun, Ling Zhao, Language Feature Mining for Music Emotion Classification via Supervised Learning from Lyrics international symposium on advances in computation and intelligence. pp. 426- 435 ,(2008) , 10.1007/978-3-540-92137-0_47
Thierry Bertin-Mahieux, Paul Lamere, Daniel P. W. Ellis, Brian Whitman, THE MILLION SONG DATASET international symposium/conference on music information retrieval. pp. 591- 596 ,(2011) , 10.7916/D8NZ8J07
Jacquelin A. Speck, Youngmoo E. Kim, Brandon G. Morton, Erik M. Schmidt, A COMPARATIVE STUDY OF COLLABORATIVE VS. TRADITIONAL MUSICAL MOOD ANNOTATION international symposium/conference on music information retrieval. pp. 549- 554 ,(2011)
Kate Hevner, Experimental studies of the elements of expression in music American Journal of Psychology. ,vol. 48, pp. 246- 268 ,(1936) , 10.2307/1415746
Shuo Chen, Josh L. Moore, Douglas Turnbull, Thorsten Joachims, Playlist prediction via metric embedding Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '12. pp. 714- 722 ,(2012) , 10.1145/2339530.2339643
Jin Ha Lee, Xiao Hu, Generating ground truth for music mood classification using mechanical turk acm/ieee joint conference on digital libraries. pp. 129- 138 ,(2012) , 10.1145/2232817.2232842