Towards Examining Supervised and Unsupervised Learning for IoT Attack Detection

Abstract

The Internet of Things (IoT) is the term used to describe the numerous physical objects/devices connected to the Internet and collecting and exchanging data globally. IoT devices are especially susceptible to network attacks, including but not limited to botnet attacks, spoofing attacks, and denial of service attacks. This thesis explores supervised and unsupervised learning approaches to compare two types of traffic flow exporters on different publicly available datasets. Evaluations and results show that it is possible to achieve high weighted average F1-scores for attack detection using off-the-shelf supervised learning algorithms and traffic flow features on IoT networks.