Network Intrusion Detection Systems (IDSs) have been widely used for monitoring and managing the network connections, and
prevent the unauthorized connections. Machine learning models have been utilized for classifying the connections into normal
connections or attack connections based on the behavior of the users. One of the most common issues face the IDSs is the low
classification accuracy of the detection system. As well as high dimensionality in the feature selection process. However, the
feature selection methods usually used for decreasing the redundancy in the datasets, and enhance the classification
performance. In this paper, a Chaotic Salp Swarm Algorithm (CSSA) was integrated with the Extreme Learning Machine (ELM)
classifier to select the most relevant subset of features and decrease the dimensionality of the dataset. Each Salp in the
population was represented in a binary form, where 1 represented a selected feature, while 0 represented a removed feature.
The proposed feature selection algorithm was evaluated based on NSL-KDD dataset, which consists of 41 features. The
experiments' outcome showed that the proposed algorithm achieved good results in terms of the classification accuracy
(97.814%) and number of selected features.