Research On Emotion Anthropomorphic Strategies In Human Computer Interaction

Since communication between human are natural and emotional,people naturally expect computer to have the ability of cognitive emotion in the interaction between humans and computers.How to make the computer recognize,understand and generate anthropomorphic emotion has received more and more attention by computer science,psychology and other disciplines,and has gradually formed the cross field of cognitive emotion computing.In order to enhance the cognitive and cognitive analysis ability of robot,this paper carries out the research work on the emotion cognition of robot in open and closed domains,and the main research contents are as follows:To solve the problems of the existing in the process of human-computer interaction system,such as lack of emotion and low participation,A cognitive emotion interaction model based on game theory in PAD(Pleasure-Arousal-Dominance)emotion space was proposed.Firstly,the interactive input emotion of participant was evaluated and some influence factors such as friendship and resonance were extracted to analyze the current human-computer interaction relationship.Secondly,modeling the emotional generation process of participants and robots by simulating the psychological game process in interpersonal communication,and the optimal emotional strategy of the robot was obtained by using the sub-game perfection equilibrium of the embedded game.Finally,the emotional state transition probability of the robot was updated according the optimal emotional strategy.The spatial coordinates of the six basic emotional states were used as labels to obtain the PAD spatial coordinate of the robot emotional state after emotional stimulate,The results of experiment show that compared with the others emotional interaction model,the proposed model can reduce the dependence of robots on external emotional stimuli and effective guide participants to participate in human-computer interaction,which provides some ideas for the emotion cognition model of robot in human-computer interaction.In the closed domain human-computer interaction system,most of the existing research focused on the response content's quality and ignored the emotion cognition of robot,lead to poor consultation experience for participants.In order to solve this problem,a cognitive emotion interaction model for customer service robot was proposed.Firstly,establishing sentiment analysis model based LSTM(Long Short-Term Memory)toregognition interaction input text.Then,the interactive input text of participant was matched with knowledge base,and the answer with higher semantic confidence was returned as candidate answer set of robot.The dialogue text of participant that may be caused by each candidate answer of the robot was predicted according the probability of interaction condition probability.Finally,the reward function was defined to evaluat the candidate answer of robot,and selecting the answer with the highest reward expectation as the optimal session strategy for the customer service robot.The experimental results show that compared with the others interaction model,the proposed model can effrctive enhance the acceptance of the customer service robot on the basis of ensuring the accuracy of the answer questions,thus promoting harmonious human-computer interaction experience.