Transparency, a principle advocated by the General Data Protection Regulation, is usually defined in terms of properties such as availability, auditability and accountability and for this reason it is not straightforwardly measurable. In requirement engineering, measuring a quality is usually implemented by defining a set of metrics for its composing properties, but conventional approaches offer little help to achieve this task for transparency. We therefore review requirements for availability, auditability and accountability and, with the help of a meta-model used to describe non-functional properties, we discuss and advance a set of metrics for them. What emerges from this study is a better justified and comprehensive tool which we apply to measure the level of transparency in medical data-sharing systems.
Spagnuelo, D., Bartolini, C., & Lenzini, G. (2017). Modelling metrics for transparency in medical systems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10442 LNCS, pp. 81–95). Springer Verlag. https://doi.org/10.1007/978-3-319-64483-7_6