A Hierarchical k-Anonymous Technique of Graphlet Structural Perception in Social Network Publishing

The structural information of social network data plays an important role in many fields of research. Therefore, privacy-preserving social network publication methods should preserve more structural information, such as the higher-order organizational structure of complex networks (graphlets/motifs). Therefore, how to preserve the graphlet structure information in a social network as much as possible becomes a key problem in social network privacy protection. In this paper, to address the problem of excessive loss of graphlet structural information in the privacy process of published social network data, we proposed a technique of hierarchical k-anonymity for graphlet structural perception. The method considers the degree of social network nodes according to the characteristics of the power-law distribution. The nodes are divided according to the degrees, and the method analyzes the graphlet structural features of the graph in the privacy process and adjusts the privacy-processing strategies of the edges according to the graphlet structural features. This is done, in order to meet the privacy requirement while protecting the graphical structural information in the social network and, improving the utility of the data. This paper uses two real public data sets, WebKB and Cora, and conducted experiments and evaluations. Finally, the experimental results show that the method proposed in this paper can concurrently provide the same privacy protection intensity, better maintain the social network’s structural information and improve the data’s utility.

[1]  K. Liu,et al.  Towards identity anonymization on graphs , 2008, SIGMOD Conference.

[2]  Lei Zou,et al.  K-Automorphism: A General Framework For Privacy Preserving Network Publication , 2009, Proc. VLDB Endow..

[3]  John Scott.  What is social network analysis , 2010 .

[4]  Lise Getoor,et al.  Preserving the Privacy of Sensitive Relationships in Graph Data , 2007, PinKDD.

[5]  Angsheng Li,et al.  Structural Information and Dynamical Complexity of Networks , 2016, IEEE Transactions on Information Theory.

[6]  Michalis Faloutsos,et al.  On power-law relationships of the Internet topology , 1999, SIGCOMM '99.

[7]  Ting Yu,et al.  Anonymizing bipartite graph data using safe groupings , 2008, Proc. VLDB Endow..

[8]  Noga Alon,et al.  Finding and counting given length cycles , 1997, Algorithmica.

[9]  Siddharth Srivastava,et al.  Anonymizing Social Networks , 2007 .

[10]  Donald F. Towsley,et al.  Resisting structural re-identification in anonymized social networks , 2008, The VLDB Journal.

[11]  Jure Leskovec,et al.  Higher-order organization of complex networks , 2016, Science.

[12]  Xiaowei Ying,et al.  Comparisons of randomization and K-degree anonymization schemes for privacy preserving social network publishing , 2009, SNA-KDD '09.

[13]  Tamir Tassa,et al.  Anonymization of Centralized and Distributed Social Networks by Sequential Clustering , 2013, IEEE Transactions on Knowledge and Data Engineering.

[14]  Chris Clifton,et al.  A Guide to Differential Privacy Theory in Social Network Analysis , 2012, 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining.

[15]  Balachander Krishnamurthy,et al.  Class-based graph anonymization for social network data , 2009, Proc. VLDB Endow..

[16]  Danfeng Yao,et al.  The union-split algorithm and cluster-based anonymization of social networks , 2009, ASIACCS '09.

[17]  Alina Campan,et al.  Data and Structural k-Anonymity in Social Networks , 2009, PinKDD.

[18]  Yanghua Xiao,et al.  k-symmetry model for identity anonymization in social networks , 2010, EDBT '10.

[19]  Jordi Herrera-Joancomartí,et al.  k-Degree anonymity and edge selection: improving data utility in large networks , 2017, Knowledge and Information Systems.

[20]  Jian Pei,et al.  Preserving Privacy in Social Networks Against Neighborhood Attacks , 2008, 2008 IEEE 24th International Conference on Data Engineering.