A feature selection algorithm for intrusion detection system based on the enhanced heuristic optimizer

With the rapid development of network technology, the dramatic growth of network traffic has also led to a large number of irrelevant features and noise, which affect the performance of network intrusion detection systems . Feature selection has thus become a key aspect in building these systems. In this paper, an enhanced heuristic optimization algorithm (EHO) is proposed, demonstrating excellent global convergence and superior search capabilities. The CEC standard test functions are used to evaluate the effectiveness of the algorithm. Experimental results show that the proposed algorithm has a faster convergence speed and stronger exploration ability when dealing with multimodal problems, significantly outperforming CSA, CSO, EFA, BWO , and RIME methods. Additionally, a wrapper feature selection method based on the optimization algorithm is proposed, and the algorithm’s performance is evaluated using three public datasets (NSL_KDD, UNSW_NB15, and CIC-IDS2018). The results indicate that the proposed method outperforms existing feature selection algorithms in terms of accuracy, precision, recall, and F1-score, achieving 90.95%, 93.39%, and 98.50% accuracy on the NSL_KDD, UNSW_NB15, and CIC-IDS2018 datasets, respectively.