Data-Driven Approach to Network Intrusion Detection System Using Modified Artificial Bee Colony Algorithm for Nature-Inspired Cybersecurity

With ever-evolving cyberspace, adaptive defense is crucial. In this paper, we show the Adaptive Defense Mechanism to identify Anomalous hosts in a network using the Artificial Bees Colonization Algorithm. A self-driven metric has been defined to determine the performance of a network that would detect the behavior of its nodes. This algorithmic metric is inspired by the Nature-Inspired Artificial Bees Colonization Algorithm. The end result is randomly generated using a dimension index that gives the same result on the node’s behavior which is then used to determine the probabilistic parametric fitness of the individual nodes. This helps to determine which nodes are getting affected the most or are nearer to the attack surface. The defense mechanism is based on the Nature Inspired Artificial Bees Colonization Algorithm, which is able to detect the nearest point/s of attack on nodes based on the experimental simulation of attacked nodes. It also shows the impact of the defense mechanism on the various topologies of the nodes as predefined in the testbed implementing a Distributed Denial-of-Service attack on the nodes. The proposed algorithm showcases the nodes that are affected due to the attack, providing the nearest point of the breach, which can provide a comprehensive way of examining the intrusion point. This algorithm outperformed in terms of stability and early identification of the malicious nodes.