A utility preserving data-oriented anonymization method based on data ordering
Mostafa Salari, Saeed Jalili, Reza Mortazavi · 2014
Due to recent advances, data collection and publishing for scientific purposes are made by some organizations. Published data should be anonymized such that being useful while privacy of data respondents are preserved. So, there is a trade-off between data utility and privacy. Microaggregation is a popular family of anonymization methods that operates on numerical data. In this paper, we propose a microaggregation algorithm called NFPN_MHM that first sorts data in a spiral shape, next it finds a partitioning with the lowest utility loss with respect to the sorted data. Experimental results show that the proposed method attains lower information loss than traditional microaggregation methods and provides a better trade-off between data utility and privacy, especially for scattered data.