| Abstract: | ![]()
In data publishing, anonymization techniques have been designed to provide privacy protection. Anatomy is an important techniques for privacy preserving in data publication and attracts considerable attention in the literature. However, anatomy is fragile under background knowledge attack and the presence attack. In addition, anatomy can only be applied into limited applications. To overcome these drawbacks, we propose an improved version of anatomy: permutation anonymization, a new anonymization technique that is more effective than anatomy in privacy protection, and in the meanwhile is able to retain significantly more information in the microdata. We present the detail of the technique and build the underlying theory of the technique. Extensive experiments on real data are conducted, showing that our technique allows highly effective data analysis, while offering strong privacy guarantees. |
