Privacy preservation in social networks through alpha — Anonymization techniques

Abstract

We propose an (a, k) anonymity model based on the eigenvector centrality value of the nodes present in the raw graph and further extend it to propose (a, l) diversity model and recursive (a, c, l) diversity model which can handle the protection of the sensitive attributes associated with a particular actor. For anonymization purpose, we applied noise node addition technique to generate the anonymized graphs so that the structural property of the raw graph is preserved. Our proposed methods add noise nodes with very minimal social importance. We applied eigenvector centrality concept over traditional degree centrality concept to prevent mixing of highly influential nodes with less influential nodes in the equivalence groups.