Design of emotional Intelligent Tutor System based on HMM

Successful ITS (Intelligent Tutor System) design depends on a large number of factors. This paper focuses on emotional interaction in ITS. An artificial emotional model based on HMM(Hidden Markov Model) is proposed. In this model, the teacher´s mood and emotion is created in the form of probability, as well as, the recognition results of students´ facial expression are regarded as the actuating signal of the artificial emotional model. Teacher´s affection is considered as a two-layer random signal, which is able to establish the emotional model of various character traits by adjusting the initial parameters. In addition, the learning evaluation model based on Bayes network is adopted to test the student´s cognitive ability. In summary, the emotional ITS based on HMM enables to achieve the emotional interaction in the procedure of the cognition.