In such practices, AI systems often manifest themselves in the form of black-boxes, providing output based on historical data analyses, but without any form of motivation, explanation, or tracability of their decisions and suggestions. The lack of explanation and tracability of decisions hinders further adoption by its users, and diminishes their potential value, as the lack of such abilities makes it hard, if not impossible, to identify and assign responsibility to the (human or AI) actors involved in the decision-making process. The main problem addressed in this work package is therefore to design and develop a computational framework that allows stakeholders to understand complex decision-making processes, determine responsibility and accountability of the decisions made, and explain those decisions in a personalized manner.