Construction and analysis of vocal music evaluation system based on BP-RBF-SVM integrated neural network

Abstract

With the development of neural network technology, integrated neural networks with better performance and stronger generalization capabilities have emerged. In order to solve the current deficiencies in the process of vocal music teaching quality evaluation in normal colleges and universities, a vocal music teaching evaluation system is proposed based on integrated neural network. The system first collects teaching evaluation data of vocal music students in normal universities. The analytic hierarchy process is used to retain important evaluation indicators, and some pre-processing of the data is used to construct evaluation learning samples; then the BP-RBF-SVM integrated neural network is used to evaluate the quality of vocal music teaching, and the evaluation results of vocal music teaching quality are obtained; finally for verification. The reliability of the model is compared with the evaluation results of a separate RBF neural network and BP neural network. The results show that the vocal music teaching quality evaluation of the integrated neural network system has a good fit and small error, which provides a new reference for the evaluation of vocal music professional teaching quality.