A Personalized k-Anonymity Privacy Preserving Method

摘要

In recent yeaxs, many data sets are collected by organizations for the purposes of research, cooperation and e-business. Releasing data set about individuals without revealing their sensitive information has become an active issue, and k-anonymity-based models are effective techniques that prevent linking attack. Few of the previous works, however, pay more attention to the issue of personalized privacy preserving and representation efficiency of anonymous data sets. Using different k parameter for each individual group and personal generalization paths represent different privacy preserving requirements of data subjects, the definition of derived set of anonymous table was proposed based on partial order over generalization/specialization of QI attributes in generalization tables. Furthermore, PK-anonymity, a personalized k-anonymity privacy preserving algorithm, was proposed. Theoretical analysis and experimental results testify the effectiveness and efficiency.