Machine Learning-Based Threat Identification Systems

The increasing popularity of internet of things (IoT), dissimilar networks, distributed devices, and applications has turned out to be a major call for the identification of novel security threats and tracing malicious network behaviours. An intrusion detection system (IDS) is a self-defense tool for preventing several types of cyberattacks. Latest machine learning (ML) methods are becoming the backbone for constructing intelligent IDS that are highly data driven. This chapter proposes decision tree based IDS for the dataset NSL-KDD. A novel approach has been developed for the ranking of security features. The proposed system has been validated against performance evaluation metrics consisting of recall, precision, accuracy, and F score. The results produced by the proposed system are compared with well-known ML methods including logistic regression, support vector machines and K-nearest neighbor in order to analyse the efficiency.