The relationship between music and human emotions is a very interesting area. But how exactly music affects human emotions and what the valid parameters needed to classify the pattern is also an open question. In this paper, we consider the effect of four different types of music on human emotions through the nonlinear analysis of Heart rate variability (HRV). The t-test shows that the Largest lyapunov exponent (LLE) of HRV acquired from subjects with or without music stimulation is

more »

... y different (p<0.01, Cohen's d>0.8). More importantly, the LLE of HRV is also influenced by different types of music respectively, and then we applied LLE as the affective features to recognize the subjects' emotional state which induced by specific types of music. An overall correct rate of 84 percent for quinary classification of amusement, excitement, sadness, fear and the baseline state by SVM classifier are obtained.