A Research On Social Network Privacy Protection Method

People in the social network of relationships and activities can be others known andshared, due to the rapid growth of the number of services and applications bring the privacyleak problem that should not be ignored in social network. Since the data in graph have veryrich information, contrast to the traditional database or relational model with privacyprotection, the privacy protection technique of released social network is still in its earlystage.Through the research of social network at home and abroad present situation of thestudy, and related modeling and technical theory knowledge, this thesis focused on thefollowing research.This thesis in view of the problem of the social network will leak user’s privacyinformation, through the research of social network’s form and characteristics, according tothe knowledge of the graph theory to model social network graph. Firstly utilizing simpleanonymous concept to prepossess the social network diagram, and then using knowledge ofboth privacy protection and graph theory to theoretically study diagram in order to getanonymous social network data restrictive conditions, the graph has been protected privacyin accordance with the condition of social network diagram. According to the different sizesof social network data, the choice of disturbance levels privacy protection, so that willprotect the social network figure after releasing out and, to some extent, it can ensure abetter protection of the user privacy, at the same time, the data anonymous have very goodutility.Finally, this thesis verified the proposed algorithm through the experiments, and theexperimental results were compared for analysis, the results show that the superiority of thealgorithm, and discuss the direction of further improvement.