Analysis of Cultural Group Communication Behavior based on Deep Belief Network Algorithm

Abstract

The role of group communication in the study of cultural group behavior is very important, but there is a problem of large research error. Information statistics cannot solve the communication problem in the study of cultural group behavior, and the behavior recognition rate is low. Therefore, this paper proposes a deep belief network algorithm for the analysis of cultural group behavior communication. Firstly, the belief network theory is used to study the communication behavior, and in-depth mining is carried out according to group communication requirements to reduce the irrelevant factors in communication. Then, the deep belief network algorithm is used to continuously divide the behavior of cultural groups and form the final behavior recognition set. MATLAB simulation shows that the deep belief network algorithm's behavior recognition accuracy and behavior recognition time are better than the information statistics method when the communication requirements are known.