Predictive models for effective policy making against university dropout

The mere development of a software to predict University dropout is not sufficient for its effective implementation in the academic context. In order to exploit it as a tool supporting decision-making, such a software should be provided with information necessary for its integration into the decision-making processes of University governance. In this work, we present a predictive tool, at the state of the art, providing a functional description of its integration through practical examples. In addition, we propose a simplified scheme to guide the reasoning on the software, which we structure according to the following processes: the learning of the machine, the choice of the representation of dropout and the interpretation of the results. Finally, we share some considerations addressing education designers and institutional decision-makers on the management of interventions inspired by the prediction of freshmen academic outcome.