Randomized Perturbation for Privacy-preserving Social Network Data Publishing

With the popularity of social networks, privacy issues regarding publishing social network data have attracted an great deal of attention from academia. We proposed a randomized perturbation method to alleviate the privacy-disclosure problem. Our method is based on a randomized perturbation matrix, which is flexible and easy to extend. We solved the problem of calculating the privacy risk of the proposed model. We used three real data sets to evaluate the privacy-preserving method. The experimental results show that our method is better in terms of data utility than the method of addition/deletion of edges.