Feature selection and classification algorithm based on support vector machine in music emotion analysis

The goal of this article is to deeply study a Music emotion (ME) classification method based on Support Vector Machine (SVM). It aims to improve the accuracy and efficiency of ME recognition. In this article, the importance of ME analysis and its current challenges are discussed, especially the problems of inappropriate feature selection and insufficient efficiency of classification algorithm. To solve these problems, this study proposes an optimized feature selection method and a new SVM classification strategy. In terms of methods, this article first creates an efficient feature selection mechanism. By analyzing the correlation between features and the role of SVM in classification, this mechanism screened out the most differentiated music feature set. Based on these selected features, a ME classification system of SVM is constructed, and through the training and prediction of the model, the accurate recognition of ME is realized. The results show that the algorithm proposed in this article achieves high accuracy, recall and F1 value in multiple emotion categories. The optimized feature selection technology and improved classification effect have made important contributions to the field of ME analysis.