Research On Privacy Protection Of Social Networks Based On Evolutionary Game

In recent years,the rapid development of social networks has diversified the ways of communication between people and gradually become an indispensable communication channel in people's work and study life.In this process,a large amount of personal information is also generated and stored.This information reveals many details about users,such as social connections,dates of birth,id numbers and so on.Therefore,more and more experts and scholars focus on how to protect the security of information and privacy generated in social networks.Before users use an App,they often need to fill in a questionnaire about their personal information,which allows service providers to discover some habits and preferences of users and provide them with a more personalized experience.On this premise,many experts and scholars in the industry put forward some targeted methods for information security and privacy protection of social networks.The core of the methods is to obtain users' privacy on the premise of ensuring service quality.The combination of information entropy and evolutionary game can not only quantify information and privacy,but also concretize abstract concepts,which is of great significance to guide subsequent research on privacy protection.Research on privacy protection in social networks is mainly based on anonymous processing,access control,differential privacy and some methods based on game theory.The first three are biased to study from the perspective of users or service providers.Game theory emphasizes that from the perspective of a third party,both parties engage in a cycle of constant game,adjustment and game,and eventually find a dynamic equilibrium state.However,the evolutionary game does not require participants to be completely rational.It is based on Darwinian biological evolution and starts from the whole life cycle of biological groups.Behavioral changes in groups are seen as internal adjustments in a dynamic system.The relationship between individual behavior and group behavior is described in this paper,and the process from individual behavior to group behavior and related factors are incorporated into the evolutionary game model to form a macro model with micro basis,which provides a theoretical basis for macrocontrol of group behavior.Based on this model allows users and service providers to connect each other needs,This paper ensure degree of privacy exposed,avoid the user with too little information to obtain accurate service,at the same time avoid additional services a large number of users' privacy and to take a profit,or harmful information security behavior,the main research content is as follows:In terms of information and privacy measures,this paper uses information entropy formula to the quantitative information and improves the information entropy formula to quantify privacy.The win-win relationship between users and service providers is regarded as a process of continuous evolution and adaptation.According to this idea,a hybrid strategy is proposed to help users judge whether the service provider is malicious or hold too much information.Given the real interaction in the App,the game only happens between the user and the service provider,so what we are looking for is a balance between the two.Information entropy theory is used to quantify the information given by users,and sensitivity coefficient is added into the privacy calculation formula to represent the sensitivity differences of different users.On this basis,this paper uses Evolutionary game theory to solve the replication dynamic process of both users and service providers,obtains Nash equilibrium Strategy according to Lyapunov stability principle,and uses Jacobi matrix corresponding to dynamic equation to make Evolutionary Stable Strategy(ESS)judgment.Finally,200 questionnaires were collected randomly,the data were preprocessed and similarity analysis was carried out,and simulation experiments were carried out to verify the Nash equilibrium strategy and judge the stability of the strategy.