A COMPUTATIONAL INTELLIGENCE-BASED APPROACH FOR THE ANALYSIS OF THE HEALTH EXPENDITURE IN BRAZILIAN MUNICIPALITIES

This study aimed to analyze the health expenditures of Brazilian municipalities using computational intelligence, specifically through cluster analysis, to identify which municipalities have similar per capita expenditures relative to their GDP per capita. The research employs a quantitative and documentary approach, examining a sample of 117 Brazilian municipalities. These municipalities were grouped into clusters using the k-means method with the Euclidean similarity metric to identify patterns in public health spending, taking into account their economic realities. Data from the Brazilian Finance Bank (Finbra), adjusted for inflation, and cluster analysis conducted via R software were utilized. The findings revealed that disparities linked to the economic characteristics of municipalities significantly influence the correlations between health expenditures, although these correlations were less pronounced concerning primary care. By elucidating these results, the study contributes to the application of computational intelligence techniques in grouping municipalities and supports financial decision-making by municipal governments and public managers in the health sector.