Modeling Fiscal Pressure for Economic Resilience: An Intelligent Information System Approach Based on J48 Trees

Abstract

The economic resilience of a state is directly influenced by its authorities’ ability to manage fiscal pressure, a crucial factor in maintaining financial independence and macroeconomic stability. Within this framework, the present paper highlights the strategic importance of information systems in supporting fiscal decision-making and proposes an original predictive model for estimating aggregate fiscal pressure. The research employs a methodology based on decision tree regression, specifically the J48 algorithm, using Eurostat fiscal data across multiple categories (consumption, labor, property, and environmental taxes). The data was pre-processed through cleaning, interpolation of missing values, and normalization to ensure consistency and comparability across EU member states. Two categories of explanatory variables were included: historical fiscal data and macroeconomic indicators such as GDP, inflation, and unemployment. The predictive model achieved a high level of accuracy (98.62%), identifying significant nonlinear relationships and classification rules among fiscal indicators. The obtained results confirm the model's performance, revealing key connections between fiscal components and providing institutional actors with robust tools for anticipating and mitigating fiscal risks. In conclusion, the integration of information systems with advanced predictive algorithms proves essential for strengthening economic security and developing forward-looking, coherent fiscal policies.