Employing Artificial Bee Colony Algorithm for Feature Selection in Intrusion Detection System

Feature selection in Intrusion Detection System (IDS) helps in optimizing the classification process. Being an optimization problem, it is vitally important to choose the appropriate subset of features from feature space. In this paper, Artificial Bee Colony (ABC) algorithm has been used for feature selection process followed by random forest classifier applied for classification task. The proposed model is evaluated over two well-known datasets, i.e. NSL KDD and UNSW-NB15. The experimental results show that the proposed approach is able to select good feature set from both datasets using 80.83% and 88.17% accuracy. The performance of the system is also compared with the existing literature work which uses same datasets.