Research On Evolutionary Game Behavior Of Collaboration Network

Social network is one of the structural social relationships naturally formed in the evolution of human societies.Cooperation is ubiquitous in social,economic and ecological systems,which is in contradiction with the essence of human selfishness.Networked Evolutionary game provides a framework for research.Previous research has focused on well-mixed structures,and yet most individuals in real interactions have sentiments and unequal status in cooperation process and changing in time.This raises the question of whether cooperation can persist despite different sentiments and unequal status of individuals.In this paper,sentiments are included the positive and the negative,and unequal statues are the small node and the big node based the node degree.We develop a game model to study cooperative behaviors based unequal statuses and sentiments,and experimentally examine the model by numerical simulation and real networks.Surprisingly,we find that small nodes are more prone to choose positive cooperation relative to comparable big nodes on the promise of enough profits of the tacit knowledge and the excess return.Our results reveal that the unequal status is the hidden mechanism for cooperative behaviors,and provide a new prospective to investigate the evolution of cooperation in more realistic environments.Individuals with the same status or strength are more inclined to interact in the same area or field,so they are more inclined to play games for their own interests.We study and establish a game model between nodes with the same state or even strength,and use network evolutionary game to detect their cooperative behavior.Two game models with different strategies are proposed: the game model between small nodes,and the game model between large nodes.The evolutionary stable point is studied by theoretical analysis and simulation.Different time-series stochastic networks are used to simulate the interaction and game relationship between them,as well as the influence of some payoff factors.One network is composed of small nodes,the other is composed of big nodes.Experiments show that punishment can improve the cooperative behavior between big nodes.The cooperation behavior of big nodes and small nodes shows different phenomena,that is,big nodes are easier to cooperate with big nodes.The cooperation level between big nodes is higher than that between small nodes.Due to the gradual diversification of networks,many networks are directional and weighted in reality.In cooperation,there will be a dominant party with strong willingness to cooperate and strong financial support.The data of patent joint application can reflect the cooperation and interaction between innovation subjects.Based on the data of patent joint application,the patent cooperation network is constructed.The output node is the first applicant and the input node is other applicants of a patent.If the first applicant and other applicants exist simultaneously in the development of a patent,a directional contact from the first applicant to other applicants is established.Based on evolutionary game theory,a game model between input node and output node is established,and what kind of protocol or allocation scheme can promote cooperation is studied.The coevolution rules of strategy and edge weight are proposed.The proposed rules can lead to the common change of node strategy and network structure.We find that if we can give more rewards to the output nodes,give enough financial support to the input nodes,and do not need more government funding,cooperation level can be improved.Moreover,if we can adjust the weight of the connected edges between nodes according to the strategy and income of the game,we can also promote cooperation.This paper constructs an evolutionary game model on the two-layer network to study the evolution of cooperative behavior on the two-layer network.First of all,the network structure is that the upper layer is the interaction network between nodes,the lower layer is the sign network,and the sign of the edge represents the friendly and hostile relationship between nodes.The strategy space of the upper layer game is cooperation and non-cooperation,while the strategy space of the lower latyer game is positive and negative.In the process of game evolution,the cooperative behavior of the upper layer changes constantly,the attitude of the lower layer also changes,and the signs and connecting edges between nodes also change.It is found that the threshold of disappear of cooperators in the game model is relatively low.When the payoff of network nodes depends on the payoff of the upper network and the lower layer network in three to seven proportions,the cooperation level is the highest and the average payoff is the highest.If the node doesn't cooperate,we will change the positive edge to the negative edge.This kind of punishment in disguise is beneficial to cooperation.The probability of the negative edge changing into the positive edge is very small,which indicates that it is a little possiblility to change the relationship from the negative edge to the positive edge.All these can be regarded as a disguised punishment for the non-cooperator.On the basis of designing a variety of evolutionary game models,we can make full use of evolutionary game to predict the possibility of node cooperation in the next time,which is reflected in the network structure it can be used to test the possibility of edge connection in the next time.Therefore,we use evolutionary game to predict the cooperative links in time series cooperative networks,and then we can use the corresponding detection indicators to evaluate the prediction results.And we compare the result of evolutionary game with that of link prediction based on node similarity.It is found that the gap between the two methods is not very big,and the basic prediction trend is consistent.