10.57979/P0Y3RS Leonardo Vanneschi Leonardo Vanneschi https://orcid.org/0000-0003-4732-3328 NOVA Information Management School Diego Costa Pinto Diego Costa Pinto https://orcid.org/0000-0003-4418-9450 NOVA Information Management School POLEN DataHub 2023 Over-indebtedness Poverty risk Economic austerity Credit control Artificial intelligence Automated machine learning Costa, Elvira Elvira Costa NOVA Information Management School 2023-05-22 2023-06-16 2023-06-16 10.1016/J.JBUSRES.2020.10.035 10.57979/P0Y3RS/DOCJKT 10.57979/P0Y3RS/FHXNGO 630699 1138 602032 571255 text/comma-separated-values text/tab-separated-values text/comma-separated-values text/tab-separated-values 3.0 Custom terms specific to this dataset The dataset comprises a broad range of variables to understand the full picture of consumers’ financial health: family socio-demographics, total income, total expenses, employment information, as well as all credit details. The features considered for the analyses were: socio-demographic characterization (marital status, level of education completed, number of people in the household), the perceived causes for over-indebtedness (from a predetermined pool of causes), and data concerning their economic situation, including the total income and expenses of the household as well as data concerning their credits and debts (amount of the monthly installments for credit cards, housing credit, car credit, personal credit and other types of credit or debts; total monthly installment concerning all credits). Each household is represented by one record (one observation) of the dataset with many features to describe their characteristics and behavior Fundação para Ciência e a Tecnologia DSAIPA/DS/0113/2019