Research On The Subsidy Policies For Networked Voluntary Vaccination Population

Such as neural networks,the World Wide Web,and social networks in real systems,the diverse forms of representation of complex networks provide a solid foundation for inter-individual connections and interactions.How to control the spread of epidemic in the networked population is a main part of complex network research.As the main method to inhibit the spread of epidemic,vaccination is mainly divided into forced vaccination and voluntary vaccination,and forced vaccination is difficult to implement in reality.Therefore,voluntary vaccination is the most commonly used method to control the spread of epidemic in the network.However,due to the self-interest of individuals,they will adjust their behavior according to the profit obtained in the system and other local information(such as other individual behavior strategies),which leads to the difficulty of achieving social optimal under Nash equilibrium.The main challenge in the current research is how to effectively solve this vaccination dilemma.The subsidy mechanism affects the behavioral strategies of a small number of individuals,causing the remaining individual's behavioral strategies changed,and then enhances the social benefits under the Nash equilibrium.Therefore,designing a suitable subsidy mechanism is the main research problem of this thesis.Most of the subsidy target of the current work is based on the centrality of the nodes,and the particularity of different vaccination games is not considered.This thesis analyzes and designs the subsidy mechanism under evolutionary vaccination game and epidemic control game.The main contributions are listed as below:(1)This thesis considers the historical information of individual behavior strategies in the evolutionary vaccination game model,and proposes HI-RAN and HI-TAR subsidy mechanisms.The HI-RAN mechanism randomly selects individuals who did not voluntarily vaccinate in the previous season;the HI-TAR subsidy mechanism combined the degree centrality of the nodes.The simulation results in different networks show that the two subsidy mechanisms proposed can effectively limit the spread of epidemic,and the HI-TAR subsidy mechanism is more effective.The micro-analysis of the evolutionary vaccination process found that the subsidy mechanism based on historical information can improve the vaccination probability of non-hub nodes.Finally,this thesis compares several subsidy mechanisms under the evolutionary vaccination game model based on the characteristics of the spectral radius of the network,and finds that the HI-TAR subsidy mechanism can effectively inhibit the spread of epidemic under this model.(2)This thesis studies the individual vaccination behavior under the epidemic control game.The epidemic control game relies on the SIS model based on the spectral radius of the network.In this thesis,the best Nash equilibrium and the worst Nash equilibrium under the epidemic control game are analyzed.By comparing the existing LDG algorithm and the enumeration algorithm,it is found that it is difficult for LDG algorithm to estimate the worst Nash equilibrium effectively,and based on this,NI algorithm is considered to estimate the worst Nash equilibrium.This thesis analyzes the TAR subsidy mechanism in the epidemic control game and finds that the TAR subsidy strategy cannot effectively reduce the social cost under the worst Nash equilibrium.Therefore,this thesis proposes a greedy algorithm to reduce social cost under the worst Nash equilibrium.