Aplicabilidade de Algoritmos de Aprendizado de Máquina para Detecção de Intrusão e Análise de Anomalias de Rede

Abstract

This work proposed an analysis of the applicability of machine learning algorithms for the improvement of intrusion detection techniques in a cybersecurity system. This theme has attained wide relevance in the current scenario due to the fact that government structures and private entities of economic, political and social relevance around the world have had as a priority goal the digital availability of their services and assets. In this way, several studies have been proposed in the last 15 years by the academic community about the aforementioned topic in order to improve the systems that defend digital services and assets. Based on this, the scope of this work was defined and contemplated the survey and the analysis of some of these studies in order to determine the current state of intrusion detection systems on the subject of the adaptability of its own capacity for intrusion detection using machine learning models and algorithms. The conclusion attested to the evolution of intrusion detection systems, highlighting the characteristics of each study, the context and possibilities of expansion and application