Adaptive classification method of electronic music based on improved decision tree

In order to improve the accuracy of electronic music classification and shorten the classification time, this paper proposes a new adaptive classification method for electronic music based on an improved decision tree. Firstly, a robust principal component analysis method is used to remove noise from electronic music and improve the quality of electronic music. Secondly, the electronic music signal is segmented using time windows, and the classification features of electronic music are extracted using short time Fourier transform. Finally, the decision tree model is improved by using fuzzy analytic hierarchy process to calculate fuzzy association parameters. Construct an improved decision tree classification function for electronic music, calculate the optimal solution of the classification function, and complete the adaptive classification of electronic music. The test results show that the proposed method can quickly and accurately classify electronic music, with a maximum classification accuracy of 98.6%.