These tools and metrics are designed to help AI actors develop and use trustworthy AI systems and applications that respect human rights and are fair, transparent, explainable, robust, secure and safe.
Modelling Unlinkability
While there have been made several proposals to define and measure anonymity (e.g., with information theory, formal languages and logics) unlinkability has not been modelled generally and formally. In contrast to anonymity unlinkability is not restricted to persons. In fact the unlinkability of arbitrary items can be measured. In this paper we try to formalise the notion of unlinkability, give a refinement of anonymity definitions based on this formalisation and show the impact of unlinkability on anonymity. We choose information theory as a method to describe unlinkability because it allows an easy probabilistic description. As an illustration for our formalisation we describe its meaning for communication systems.
About the metric use case
You can click on the links to see the associated metric use cases
Objective(s):
