Toward a new way of minimizing the loss of information quality in the dynamic anonymization
Salaheddine Kabou, Sidi Mohamed Benslimane, Abdelbaset Kabou · 2020
In privacy preserving data publishing, the privacy models focuses only on single release of anonymzed data. The scenario of publishing is more complicated where the data is anonymized in different manners and at different times. Privacy Preserving Dynamic Data Publishing is the paradigm that addresses the dynamic data anonymization for various purposes. Improving the data utility and minimizing the loss of information quality after each release is the major issue in the dynamic anonymization context. The most operation used for the data anonymization is generalization. This operation is based on taxonomy called Value Generalization Hierarchies VGH which is responsible for the guidance of the anonymization. In this work, we aim to preserve the data utility based on the notion of ontology to quantitatively asses the quality of Value Generalization Hierarchies in terms of semantic.