Desenvolvimento de um modelo preditivo para auxiliar na identificação de lavagem de dinheiro

Abstract

This work presents the use of machine learning techniques to detect evidence of money laundering. From machine learning studies with supervised learning algorithms, a model was developed capable of identifying patterns of suspicious movements of money laundering crime in the current account of customers of a financial institution that need to be communicated to Organs competent bodies for greater investigation. The methodology developed, its results and its applicability were tested and validated in the present study, as well as suggestions for future improvements to the model, allowing such situations to be identified for decision-making by those responsible.