Implementation of Music genre classification using Support Vector Clustering algorithm and KNN Classifier for improving accuracy
M. Pavan Venkata Naga Sai, S. Kalaiarasi · 2023
To enhance the Music genre classification using novel Support Vector clustering Algorithm and K-Nearest Neighbors Classifier with improved accuracy.Two groups such as novel Support Vector Machines Algorithm and K-Nearest Neighbors Classifier are applied. The total number of samples analyzed using this methodology are 1000 music files. Among this sample dataset, 300 music files[70%] of the dataset was set to use as a training tool and 700 [30%] was taken as a testing dataset. Programming experiment was carried out for N=10 iterations for the novel Support Vector clustering Algorithm and K-Nearest Neighbors Classifier algorithm respectively. Computation processes were executed and verified for exactness. Each group consists of a sample size of 10 And the alpha value is 4.020 and the beta value is 1.359. The SPSS was used for predicting significance value of the dataset considering G-Power value as 80%.Novel Support Vector Clustering Algorithm shows a high accuracy and homogeneity for Music genre classification, and statistical significance difference is less than 0.001 (p>0.05).This research attempts to provide a new method of classifying music genres. Comparison results show that efficiency of the novel Support Vector clustering technique is superior to K-Nearest Neighbors Classifier for Music genre classification.