Abstract
Risk management has become a foundational aspect in numerous industries, propelling the implementation of machine learning technologies for impact assessment, prevention, and decision-making processes. Nevertheless, lacunae in the extant literature persist, particularly with regard to the identification of emergent trends and transversal applications. This study addresses this limitation through a bibliometric analysis of scientific production in Scopus and Web of Science, adhering to the PRISMA-2020 declaration. The findings reveal a substantial growth in publications on machine learning applied to risk management, with an increase of 98.99% between 2018 and 2023. China, South Korea, and the United States are identified as the primary research-producing countries. The analysis also identifies emerging trends, such as the application of machine learning in the evaluation of urban trees and the management of risks associated with the pandemic of severe acute respiratory syndrome (SARS-CoV-2). Key terms include random forest, support vector machines (SVM), and credit risk assessment, while terms such as prediction, postpartum depression, big data, and security emerge as new areas of study. Furthermore, there is a transition from traditional approaches such as stacking to advanced deep learning and feature selection techniques, reflecting the evolution of the discipline.