A Combinatory Feature Selection Method using Gray Wolf Optimization and Crow Search Algorithms for Intrusion Detection Systems

journal of Artificial Intelligence in Electrical Engineering

دوره 11، شماره 43، اسفند 2022، صفحه 1-10   

نوع مقاله: Original Article

نویسنده

Kayvan Asghari^*

Department of Computer Engineering, Sardroud Center, Tabriz Branch, Islamic Azad University, Tabriz, Iran

چکیده

The crow search algorithm and the grey wolf optimizer are two favored optimization approaches that have attracted extensive attention from researchers. The swarm-based lifestyle of crows in nature is the inspiration source of the crow search algorithm. The grey wolf optimizer is inspired by the hierarchical system of grey wolves for hunting. Both mentioned algorithms perform properly for solving many optimization problems, but these algorithms do not perform well for some. A combinatory optimization algorithm is introduced in this paper by combining the crow search algorithm with the grey wolf optimizer. The introduced approach has more diverse movements to explore the search space of the investigated problem. The combinatory algorithm is used to solve the feature selection problem of intrusion detection systems, where its goal is to improve the accuracy rate by selecting the most important features to build the system's classifier. The UNSW-NB15 intrusion detection dataset is considered for evaluation of the combinatory algorithm. The results of the experiments reveal the high efficiency of the combinatory algorithm for most instances in the experiments in comparison with the other popular optimization algorithms.

کلیدواژه‌ها

Gray wolf optimization؛ Crow search algorithm؛ Intrusion detection؛ Feature selection؛ Combinatory optimization algorithm