Research On Social Network Privacy Preserving Method Based On Perturbation Matrix

Social networking is an indispensable part of people in daily life; people can share information with each other. Therefore, the protection of the privacy information in the social networking can not be ignored. Social network contains a large number of users'information, the traditional methods are not very good to protect the information, so research in this area has already became a new focus.After an intensive study on the exsiting privacy preserving methods, it can be concluded that most of the methods try to find a balance between the privacy degree and data utility. Because the relationship between the privacy degree and data utility is very special, if implemented an increase of the degree of privacy protection will reduce the data availability, and vice versa. For the problem of leaking in social network, a new privacy preserving method is proposed in this paper called the Perturbation Matrix Algorithm --PMA. There are three main steps of the algorithm: firstly, the original data in the dataset is classified into a class of the same nature; secondly, the characteristics of social networks can be used in the graph to be modeled,then the operation of simple anonymize is used;finally, perturbation scheme based on knowledge of graph theory is proposed.By analyzing the degree of privacy protection, the new social network graph generated by the application of the algorithm can not only protect the users' privacy information, but also can prevent subgraph attack.At last, a series of experiments demonstrate that privacy preserving method in this paper is feasible and efficient. The direction of the further exploration is discussed.