Music Genre Classification using Linear Regression Compared with Extreme Gradient Boost Algorithm with Improved Accuracy

Abstract

Aim: To classify music based on its genre Using Machine Learning Algorithm Extreme Gradient Boosting compared with Linear Regression. Materials and Methods: The categorizing is performed by taking a Sample Size of n= 10 in Linear Regression and Sample size of n =10 in Gradient Boosting regression with the g-power value of 80% and Datasets containing recent study findings and a criterion of 0.05 percent, confidence interval of 95 % mean and standard deviation are collected from various web sources. Results: The values obtained in terms of Accuracy are Identified based on the condiment Gradient Boosting Regression (59.0%) over Novel linear Regression (53.0%). The Novel linear Regression algorithm and Extreme Gradient Boosting were found to have a statistically significant difference (p<0.05). Conclusion: The classification of music based on genre using Gradient Boosting Regression appears to be more accurate when compared to Linear Regression.