Classification of Pop and RnB (Rhythm and Blues) Songs with MFCC Feature Extraction and K-NN Classifier
Abstract
Classification is a technique for designing functions based on observations of attributes in a data so that data can be mapped that do not have a class which in this study can be called genres, into data that has been classified according to the given rules. In this research, music classification is conducted to determine whether the class or genre of music is pop or RnB (Rhythm and Blues) by using MFCC as the feature extraction method and K-NN as the classification method. The test results in this study obtained an accuracy of 77.5% with an optimal value of k = 31 as a parameter in K-NN.