Towards internal privacy and flexible K-anonymity

The focus on k-anonymity enhancements along the last decade, allows this method to be elected as the point of start for many anonymization researches. In this paper we propose « Contra-Loss », a novel anonymization approach applied at internal level, to enhance local privacy where data is at rest and « Flexible k- anonymity » which aims to apply k-anonymity in most situations by defining a semantic ontology which distinguishes between scanty and abundant quasi-identifiers to achieve adequate k-blocks.