Challenges in Cybersecurity and Machine Learning

Abstract

Cybersecurity challenges are increasing in complexity and intensity. There is a need for cybersecurity to evolve and allow cyber defenders to win against hackers’ ever-changing attacks. The latest advances in artificial intelligence bring an optimistic approach to use different methods like machine learning algorithms. However, many of those methods lack accuracy and their security is not well studied. They handle huge volumes of data of different types and they can be as well vulnerable to cyber attacks. Moreover, the perpetrators are also interested in using machine learning techniques for their bad deeds. This paper entails both opportunities and challenges of machine learning in the age of cybersecurity. The motivation is for cybersecurity professionals to assess and improve tools that use machine learning, and to support research of new ontologies. © 2022, International Publications. All rights reserved.