An overview of the usage of artificial intelligence-based intrusion detection methods

Huge numbers of computer across the globe, operating on a variety of hardware or software platforms, link to provide communication and commercial services over the Internet. However, since computers are interconnected, malevolent individuals may abuse resources and launch Internet assaults. The ever-increasing number of Internet assaults makes it difficult to create flexible, adaptable security solutions. One of the most significant components used to identify Internet assaults is the intrusion detection system (IDS). Various methods from various disciplines have been used in the literature to create effective IDS. Artificial intelligence (AI)-based methods have an important part in the development of IDS as well as provide many advantages over other approaches. There is, however, no thorough evaluation of AI-based methods to evaluate and comprehend their present position in solving intrusion detection issues. Various AI-based methods have been discussed in this article, with an emphasis on IDS creation. The sources of data sets, processing criteria, method utilized, dataset, classifier design, dimension reduction technique used, and other experimental environment setup have all been compared. The advantages and disadvantages of AI-based methods were addressed. The article will aid in a better understanding of the many avenues in which IDS research has been conducted. The results of this article are helpful for individuals interested in applying AI-based methods to IDS and related areas, since they offer important insights into the literature. The study also discusses the research's future directions in this field.