Research On Privacy-Preserving Algorithms Of Social Networks Based On Attribute Encryption

With the rapid development of information exchange technology in the Internet era,the wide application of social networks has emerged.Social network service system is a platform,which can provide users with more convenient,integrated and intelligent information exchange services,and establish the connection of social network between people.Because of its convenience and easy operation,social network has become more and more popular in people's life and work.However,due to the frequency and diversity of information exchange in social networks,the privacy issue has also attracted people's attention.How to find a balance between the degree of privacy protection and service accuracy is an urgent problem to be solved,the work and innovation are as follows:Firstly,a new idea is put forward in data publishing.On the basis of the original idea,the data source is converted,then noise is added and converted into interval tree,and then data is released.In the process of histogram transformation,a new method of constructing interval tree based on arbitrary structure is proposed to improve the query accuracy.This method guarantees the uniform conversion of data and improves efficiency.Secondly,a privacy protection scheme for social networks based on attribute encryption(PPSSN)is proposed,in which user rights are managed by data owner and attribute management server together,which not only reduces the overhead of data owner,but also avoids collusion attacks between attribute management server and users who have no access rights.In order to realize the balance between the usability of data released by users and the protection of information privacy,this scheme classifies users,designs and implements access control of different users with different privileges.In addition,the Friends Data Cache mechanism is introduced to improve and optimize the original scheme,reduce the cost of decryption,and optimize the overall query efficiency and system overhead.Finally,considering the characteristics of network nodes such as limited resources and time delay,a model of malicious software propagation is established to simulate its propagation in social networks based on epidemiology theory.The correctness and completeness of the state transition relations,the existence of equilibrium points and the local and global stability in the model system are analyzed and proved.