Title :
Revisiting Inter-Genre Similarity
Author :
Sturm, Bob L. ; Gouyon, Fabien
Author_Institution :
Audio Anal. Lab., Aalborg Univ. Copenhagen, Copenahgen, Denmark
Abstract :
We revisit the idea of “inter-genre similarity” (IGS) for machine learning in general, and music genre recognition in particular. We show analytically that the probability of error for IGS is higher than naive Bayes classification with zero-one loss (NB). We show empirically that IGS does not perform well, even for data that satisfies all its assumptions.
Keywords :
Bayes methods; error statistics; learning (artificial intelligence); music; pattern classification; IGS; error probability; intergenre similarity; machine learning; music genre recognition; naive Bayes classification; zero-one loss; Abstracts; Indexes; Materials; Niobium; Pattern recognition; Training; Vectors; Content-based processing and music information retrieval; pattern recognition and classification;
Journal_Title :
Signal Processing Letters, IEEE
DOI :
10.1109/LSP.2013.2280031
