Building improved gender equality composite indicators by object-oriented Bayesian networks

This work proposes a novel methodology for constructing gender equality indicators using an Object-Oriented Bayesian Network (OOBN). The methodology is illustrated by focusing on the composite indicator known as Gender Equality Index, annually released by the European Institute of Gender Equality (EIGE). By using province-level ISTAT data, the index is re-constructed in a modern AI environment, able to enhance its information capacity and, at the same time, to preserve its original architecture. The modularity of the OOBN ensures a computational logic that is consistent with composite indicators, while also providing additional information about the relational structure of variables.