Emotional classification based on the tempo and mutation degrees

The article put forward a model for music emotion classification research. The main research contents are extracting the tempo from the MIDI file and defining the rhythm mutation degree. Let the two factors as the input of the music emotion model, we simulated the model in matlab. We used the improved support vector machine (SVM) classification algorithm to classify the emotion. In order to achieve the precise classification, we trained the SVM by a variety of samples.