Mapeamento bibliométrico da evolução do machine learning na economia

Abstract

This paper maps the scientific production on the application of Machine Learning (ML) in Economics through a bibliometric analysis. The study aims to identify the intellectual and social structure of this emerging research field. Using the Bibliometrix package from the R software, the descriptive research analyzed 913 unique articles extracted from the Scopus and Web of Science databases, published between 1996 and 2024. The results confirm an exponential growth in publications, with an annual growth rate of 21.42%, particularly concentrated after 2017. The thematic analysis revealed a field structured around a methodological core that feeds into two main fronts: one consoli dated in Quantitative Finance and another, emerging, in the Economics of Sustainabi lity. Additionally, the study identified a leadership in publications from institutions in the United States and a still fragmented global collaboration network.