Classification of music by composer using fuzzy min-max neural networks
Pasha Sadeghian, Casey Wilson, Stephen Goeddel, Aspen Olmsted · 2017
This work utilizes high-level musical features extracted from a large music database of Sonata pieces composed by Beethoven, Corelli, and Mozart, and assesses the accuracy of Fuzzy Min-Max (FMM) Neural Network and Enhanced Fuzzy Min-Max (EFMM) Neural Network classifiers in classifying the classical pieces by composer. Results of the assessment are provided and show different accuracies depending on the parameters used in the FMM and EFMM models. This study presents a novel approach to the classification of music by composer by presenting two classifiers, namely FMM and EFMM Neural Networks, capable of classifying classical music by composer.