Analysis of Occupational Profiles in the Brazilian Workforce Based on Non-Negative Matrix Factorization

In this work we analyse occupational profiles in the Brazilian workforce through data analysis. To achieve that, we implement NMF clustering over data provided by the Brazilian Ministry of Economy. Additionally, we make use of the data on probability of automation of occupations published in the work of Frey and Osborne, originally developed in the context of the US workforce, to extrapolate a model for the Brazilian workforce. Finally we analyze the resulting datasets, focusing on investigating in which ways these combined data could inform us about the occupational profiles in Brazil.