Feature Selection Model Based on Gorilla Troops Optimizer for Intrusion Detection Systems

Abstract

Cyber security is a fundamental challenge to the Internet of things (IoT) and smart home environments.This paper presents a modified method to ystem (IDS).setection dntrusion ienhance the performance of the This modification is achieved by introducing an alternative feature selection (FS). ptimizer (GTO) algorithm.oroops torilla gmodel based on the Recently, FS has played a significant role in increasing the detection of anomalies in IDSs. To evaluate the efficiency of the developed method, a set of experimental conducted using three datasets, including NSL-KDD, CICIDS2017, and Bot-IoT datasets.asresults w xtraction (FE) model to reduce the dimensions of these datasets as a first step.Teeature f used as a areetworks (CNN) neural nonvolutional cThe hen, the extracted features are passed to the FS model for detection. The results of the developed method are compared with the well-known IDS technique. The results show the superiority of the developed method over all other methods according to the performance metrics.