Security Characterization and Expression in Data Publishing Using Selection Algorithm

Abstract

The expanding enthusiasm for gathering and distributing a lot of people information open for various purposes to research, showcase examination and practical measures have made significant security worries about people delicate data. Privacy-preservation data publishing has received lot of thoughtfulness, as it is always a problem of how to protect database of high dimension. We propose a structural importance aware approach to quantify the vulnerability/de-anonymizability of graph data to structure-based De-Anonymization (DA) attacks. We quantify both the seed-based and the seed-free Relative De-anonymizability (RD) of graph data for both perfect DA  and partial DA under a general data model trust evaluation mechanisms among different entities in human society are fitted and the multi-granularity selection standard of trust levels based on Gaussian cloud transformation is constructed. They are non-sensitive data and sensitive data.