A User-Demand Privacy Preserving Framework Based On Association Rules And Differential Privacy In Social Networks

With the rapid development of the mobile Internet,the wide range and diversity of its application scenarios have brought many conveniences to people's work and life,however many other problems have arisen,such as user information security and privacy leakage.In an insecure network scenario,an attacker can obtain a search record of the user's location,requirements,and other information through technical means,and dig out the user's private information through means such as big data analysis.At the same time,the attack can associate the private information with the user.Once an attacker attacks the user through the obtained data information,it will lead to serious consequences such as user property loss,reputation crisis,and even threat to life safety.It can be seen that privacy leakage may lead to serious personal safety hazards and even harm the society.As people's awareness of privacy leaks and the need for privacy protection continues to increase,there are urgent need for effective methods or means to achieve privacy protection.So far,research based on location privacy services has yielded more results and can effectively protect users' location privacy.However,there are few research results on demand privacy protection,and the demand privacy protection system needs to be improved.In order to improve the demand privacy protection system and reduce the risk of disclosure of users' demand privacy,this dissertation aims to design a demand privacy protection system in social networks and realize the effective protection of users' demand privacy under a certain quality of service(QoS)based on the existing privacy protection technology.First,this dissertation will study the privacy protection of users' single demand and establish a system model.Through the combination of association rules and differential privacy,the purpose of effectively protecting the privacy of users' single demand is achieved.In order to solve the relationship between user's privacy level and QoS,this dissertation will use the non-zero sum game and zero-sum game solving method in game theory.A dynamic demand privacy protection algorithm is designed to meet the different privacy preferences of users to improve the model.The algorithm can adjust the parameters according to the privacy level required by the user to meet the privacy pro tection requirements of users.On the basis of the user's single demand privacy protection,the user's demand track privacy protection is studied.It is mainly divided into the past-current demands privacy protection model and the current-future demands privacy protection model,and detailed analysis is carried out for each scenario.By incorporating differential privacy into the confidence level,the purpose of protecting the privacy of the user's demand track is achieved,and the effectiveness of differential privacy is verified.Finally,the game theory method is still used to solve the relationship between privacy level and QoS,and the requirements of user privacy protection are met under the requirement of user QoS.In summary,this dissertation plays a certain role in protecting the privacy of users in social networks,and has certain theoretical and practical value.