ABSTRACT
The goal of this paper is to investigate intelligent classification methods for the Latin musical genres classification task from a real data set. In this study, the predominant emotional characteristics of each genre are also considered, with the aim of enabling the user to choose music according to the required emotion. The proposed methods are based on the Bayesian classification method, which uses the algorithm BayesRule, able to investigate data probabilistic uncertain, and on the fuzzy classification method, whose parameters of the functions are fitted by a neuro-fuzzy system, which allows to evaluate mergers between different musical genres. These methodologies extract linguistic classification rules, which allow comparing results and performing an intelligent classification of the data by taking into account uncertainties and fusion of musical genres.
Keywords:
musical genres; Bayesian classification; fuzzy classification