Music Category Based on Adaptive Mutation Particle Swarm Optimization BP Neural Network

Abstract

To overcome the inadequacies of the traditional classification methods, an adaptive mutation particle swarm optimization BP neural network was introduced in the music category. Adaptive mutation particle swarm optimization algorithmhas the advantages of PSO and GM algorithm. Using the algorithm to find a better network weights and threshold can not only overcome the slow convergence of basic and easy to fall into local minimum limitations, but also has a high accuracy of the model. The simulation results show that the algorithm has higher accuracy than traditional classification methods, and verified adaptive mutation particle swarm optimization is an effective classification to optimize BP neural network.