Preserving network privacy with a hierarchical structure approach

With the development of Internet technologies, the influences on people´s lives of online social networks (OSN) gradually increase. These OSNs often contain sensitive information, such as the social contacts in the email, the personal consumption records in the e-commerce and so on. Therefore, the disclosure of such information would lead to violation of personal privacy. In this paper, we propose a differential privacy approach based on a hierarchical network model. The OSN structure is obtained by the hierarchical random graph (HRG) model. The differential privacy is guaranteed by a Markov chain Monte Carlo (MCMC) sample method. MCMC method ensures the utility of network data. The results of the experiment on two real world datasets show that our approach can effectively protect the critical information on the network while maintaining a good data utility.