Anonymization by Local Recoding in Data with Attribute Hierarchical Taxonomies

摘要

Individual privacy will be at risk if a published data set is not properly de-identified. k-anonymity is a major technique to de-identify a data set. Among a number of k-anonymisation schemes, local recoding methods are promising for minimising the distortion of a k-anonymity view. This paper addresses two major issues in local recoding k-anonymisation in attribute hierarchical taxonomies. Firstly, we define a proper distance metric to achieve local recoding generalisation with small distortion. Secondly, we propose a means to control the inconsistency of attribute domains in a generalised view by local recoding. We show experimentally that our proposed local recoding method based on the proposed distance metric produces higher quality k-anonymity tables in three quality measures than a global recoding anonymisation method, Incognito, and a multidimensional recoding anonymisation method, Multi. The proposed inconsistency handling method is able to balance distortion and consistency of a generalised view.