Privacy Threats and Privacy Preservation in Multiple Data Releases of High-Dimensional Datasets

A major challenge is when datasets are released to be utilized in the outside scope of data-collecting organizations, it is how to balance data utilities and data privacy. To achieve this aim in data collection (datasets), several privacy preservation models have been proposed, such as k-Anonymity and l-Diversity. Unfortunately, these privacy preservation models can be sufficient to address privacy violation issues in datasets that do not have high-dimensional attributes. For this reason, a privacy preservation model, LKC-Privacy, can address privacy violation issues in high-dimensional datasets to be proposed. With this privacy preservation model, datasets cannot be concerned about privacy violation issues when all L-size distinct quasi-identifier values are distorted (suppressed or generalized) to be at least K indistinguishable tuples. Moreover, every protected sensitive value relates to each group of indistinguishable quasi-identifier values; it must have the confidence of re-identifications to be at most C. Although LKC-Privacy is more efficient and effective than k-Anonymity and l-Diversity, it is generally efficient and effective in addressing privacy violation issues in location-based datasets. Moreover, we see that datasets satisfy LKC-Privacy constraints, they still have privacy violation issues from using data comparison attacks, and they further have data utility issues that must be addressed. Therefore, a new privacy preservation model can address privacy violation issues in high-dimensional datasets, such that its released datasets are not concerned about privacy violation issues from using data comparison attacks and are highly efficient and effective in data maintenance to be proposed in this work. Furthermore, we show that the proposed model is more efficient and effective by using extensive experiments.