Leveraging the power of Machine Learning techniques to Detect Botnet Attacks in IoT-5G Networks

The growing presence of the Internet of Things (IoT) has a value accessibility to our daily lives. Nevertheless, IoT devices are also prone to various security vulnerabilities, making them attractive targets for cyberattacks, particularly botnet attacks, which are a widespread threat. Here, we propose a decision tree-based learning model that can be implemented in detecting botnet attacks. Furthermore, we propose a novel feature selection methodology that estimates mean mutual information between features in a given data and corresponding class labels. Using high mean mutual information features, this methodology enhances a decision tree by focusing on more informative attributes. The hybrid approach is established with experiments on the CICIDS dataset that contained examples of botnet attacks.