The Research Of Group Cooperative Phenomenon Based On Networked Evolutionary Game

The phenomenon of group cooperation is widespread in real life.From small microbial groups to human society,the complex functions of the system are achieved through the synergy between basic components.Therefore,it is of great practical significance to study how to control group cooperative behavior in these complex systems.In this paper,the semi-tensor product of matrix and the computer simulation technology based on Monte Carlo method are used to study the mechanism of cooperation on complex networks.The feasibility of the methods is verified by experimental results.Firstly,since individual behavior selection in the process of population evolution can be regarded as a process of logical operation and the semi-tensor product method can represent the logical operation process by matrix operation.Therefore,this paper establishes the individual strategy evolution equation by using the semi-tensor product method of matrix,and further constructs the group strategy evolution model by using the individual strategy evolution equation.In addition,the game mechanism of virtual nodes is introduced to control the strategy evolution process of the group.Finally,the semi-tensor product method is used to model and analyze the examples.Besides,the feasibility of the methods is proved by the simulation example.Secondly,in the process of individual game,if the cooperators in the neighborhood could be identified in a timely manner and the individual can only play the game with these cooperators,the problem of cooperation dilemma can be well solved,because the game between the individual and the cooperator can ensure that their benefits will not be reduced.As for how to distinguish the collaborators in the neighborhood,the paper adopts the idea of constructing an intermediary organization,which is responsible for assigning cooperative labels or defector labels to each individual according to the historical strategy information of each individual.Individuals need to get the tag information of the game object firstly and then decide whether to play the game or not.In the individual recognition model,the process of assigning intermediaries label and individual game to earn profits are joined together through the process of a two-layer network structure.The information interaction between the two layers of the network respectively represents that the individual gets the neighbor’s tag information from the intermediary organization and that the intermediary organization collects the individual’s policy information for the next tag allocation.Finally,the performance of the model is verified by computer simulation from different aspects.Finally,the paper applies network evolutionary game to the information management of network community.In the evolution process of network community,we use evolutionary game model to model the interaction process between community users and information,and guide community users to actively participate in the management and maintenance of community topic information through reward mechanism,so as to effectively solve the problem of passive management in traditional network community.The simulation results show that the model is effective in building a healthy network community.