Análise de algoritmos de Machine Learning utilizando a biblioteca PyCaret para detecção de fraudes em transações online de cartão de crédito

Abstract

The analysis of machine learning algorithms for detecting fraud in online credit card transactions is a crucial aspect of financial security. Detecting fraud in online credit card transactions is challenging due to the constant evolution of fraudulent techniques. As a methodology, the PyCaret library was employed. This library facilitated experimentation with different machine learning models, such as Random Forest, Gradient Boosting, and Support Vector Machines, optimizing accuracy in the detection of fraudulent activities. Additionally, the library enabled the interpretation of results and visualizations, assisting in understanding the performance of the algorithms.