AI-BASED INTRUSION DETECTION SYSTEM IN CLOUD COMPUTING

Abstract

It is possible to communicate with others and do business across this global network, which is comprised of hundreds of millions of computers running a variety of hardware and software configurations. This makes it simpler for hackers to abuse resources and conduct Internet attacks since computers are linked to one another. There are significant roadblocks to the development of a security-oriented approach that may be flexible and adaptive in light of the expanding number of Internet assaults. An intrusion detection system is necessary for the identification of online threats (IDS). Maintaining the security of a network requires the use of an intrusion detection system or IDS. Because the cloud platform is continuously expanding and becoming more prevalent in our everyday lives, it is imperative that we develop an effective IDS for it as well. On the other hand, typical intrusion detection systems may encounter difficulties when used in the cloud. A cloud segment may get overburdened by the pre-determined IDS architecture because of the added detection overhead. In the context of a networked system with an adaptable architecture. Using a neural network-based IDS, we show how to make full utilize available resources while not putting undue strain on any single cloud server. The proposed IDS uses a neural network machine learning to identify new threats even more effectively.