A hybrid intrusion detection system based on ABC-AFS algorithm for misuse and anomaly detection

Due to the widespread use of the internet, computer systems are prone to information theft that has led to the emergence of Intrusion Detection Systems (IDSs). Various approaches such as machine learning, Bayesian-based algorithms, nature-inspired metaheuristic methods, swarm intelligent algorithms, and Markov neural networks have proposed to choose effective and efficacious features and improve the performance of intrusion detection systems. In this paper, we propose a new hybrid classification method based on Artificial Bee Colony (ABC) and Artificial Fish Swarm (AFS) algorithms. The Fuzzy C-Means Clustering (FCM) and Correlation-based Feature Selection (CFS) techniques are applied to divide the training dataset and remove the irrelevant features, respectively. In addition, If-Then rules are generated through the CART technique according to the selected features in order to distinguish the normal and anomaly records. Likewise, the proposed hybrid method is trained via the generated rules. The simulation results on NSL-KDD and UNSW-NB15 datasets demonstrate that the proposed method outperforms in terms of performance metrics and can achieve 99% detection rate and 0.01% false positive rate. In addition, analysis of computational complexity and time cost illustrate that overhead of the proposed method is comparable with counterpart approaches.