Slicing: A New Approach for Privacy Preserving Data Publishing

TLDR

Existing anonymization methods such as generalization and bucketization either lose significant information, especially in high‑dimensional data, or fail to prevent membership disclosure and are unsuitable when quasi‑identifiers and sensitive attributes are not clearly separated. The paper introduces slicing, a novel technique that partitions data both horizontally and vertically. Slicing partitions data horizontally and vertically and includes an efficient algorithm that computes sliced data satisfying ℓ‑diversity for attribute disclosure protection. Experiments demonstrate that slicing outperforms generalization and bucketization in preserving data utility, protects against membership and attribute disclosure, and remains effective on high‑dimensional data.

Abstract

Several anonymization techniques, such as generalization and bucketization, have been designed for privacy preserving microdata publishing. Recent work has shown that generalization loses considerable amount of information, especially for high-dimensional data. Bucketization, on the other hand, does not prevent membership disclosure and does not apply for data that do not have a clear separation between quasi-identifying attributes and sensitive attributes. In this paper, we present a novel technique called slicing, which partitions the data both horizontally and vertically. We show that slicing preserves better data utility than generalization and can be used for membership disclosure protection. Another important advantage of slicing is that it can handle high-dimensional data. We show how slicing can be used for attribute disclosure protection and develop an efficient algorithm for computing the sliced data that obey the ℓ-diversity requirement. Our workload experiments confirm that slicing preserves better utility than generalization and is more effective than bucketization in workloads involving the sensitive attribute. Our experiments also demonstrate that slicing can be used to prevent membership disclosure.