The emergence of new networking models, such as Cloud Computing and the Internet of Things, has created new security challenges, requiring the development of new mechanisms to ensure the integrity, availability, and security of information and service data. In order to create a secure and trustworthy cloud computing environment, intrusion detection is a critical tool. Accordingly, an intrusion detection system is needed in cloud environments to detect new and unknown attacks with high accuracy. One of the most researched strategies that meets these requirements is the use of hybrid machine learning techniques to automate the intrusion detection process. This paper describes how to implement a hybrid machine learning in a cloud computing intrusion detection system (IDS). The intrusion detection system is implemented as a network space at the cloud hypervisor level, which improves the intrusion detection system's accuracy. The system employs a hybrid algorithm that combines Kmeans clustering and SVM (Support Vector Machine) classification algorithms. The analysis shows that ML hybrid technology has higher detection accuracy than other ML technology for model development work that can detect well-known attacks.

Cloud Computing, Intrusion Detection System (IDS), Hybrid Machine Learning, K-means clustering and SVM (Super Vector Machine).