A formalization of re-identification in terms of compatible probabilities
This work addresses the problem of ensuring rigorous definitions for re-identification algorithms in data privacy, but it appears incremental as it focuses on formalization without new empirical gains.
The paper formalizes re-identification algorithms in data privacy using true probabilities and compatible belief functions to exclude those that fail to meet minimum requirements, but does not report any concrete results or numbers.
Re-identification algorithms are used in data privacy to measure disclosure risk. They model the situation in which an adversary attacks a published database by means of linking the information of this adversary with the database. In this paper we formalize this type of algorithm in terms of true probabilities and compatible belief functions. The purpose of this work is to leave aside as re-identification algorithms those algorithms that do not satisfy a minimum requirement.