Intrusion detection in computer networks using hybrid machine learning techniques

The emergence of new networking paradigms like Cloud Computing and the Internet of Things has introduced new security challenges that deserve new mechanisms to guarantee the integrity, availability and confidentiality of information and services to the users. One of the currently most studied strategies to satisfy these necessities is the use of hybrid Machine Learning techniques to automatize the process of intrusion detection in computer networks. This paper presents the design, implementation and performance analysis of multiple hybrid Machine Learning models for the task of intrusion detection in computer networks. Our results show that the combination of supervised and unsupervised learning algorithms complement each other to the task of creating a model capable of adapting to the detection of known and unknown attacks.