Recognition of Social Anxiety through the Analysis of BVP Volume Series

Social anxiety (SA) is highly prevalent in the general population, usually causing the impairment of social function and bringing mental and physical tortures to people. Accurate recognition of SA is significant to help people alleviate their anxiety through appropriate ways of emotion regulation. This paper presents the blood volume pulse (BVP) volume series for the recognition of SA. 30 subjects' high anxiety state and low anxiety state are respectively induced by impromptu speaking with 21 audience and without audience. The mean and the weighted Krawtchouk moments are extracted as the features. Two sample t-test shows that the higher degree of anxiety, the smaller mean of the BVP volume series. Support vector machines and leave-one-subject-out cross validation are used for the recognition of SA. The result shows that the recognition accuracy of high anxiety state and low anxiety state are 90% and 86.67%, respectively.