Using Bayesian Networks with Human Personality and Situation Information to Detect Emotion States from EEG

Emotional interaction is an important aspect of the interaction between humans and robots. Further, emotion affects a variety of cognitive processes and thus might leads to accidents. Finding ways to recognize emotion of humans has received a great deal of research attention. In this paper, the recognition model of multi-emotion states from electroencephalogram (EEG) is proposed based on Bayesian Networks with human personality and situation information as causes. Several kinds of emotion states were elicited with videos and EEG signals from fourteen channels were acquired. Experimental results from six subjects suggest that the proposed model have good performance, indicating the feasibility of using EEG to detect multi-emotion states.