Hybrid intrusion detection system using machine learning

Abstract

Recent technologies and innovations have encouraged users to adopt cloud-based architectures. 1,2 This has reduced IT barriers and provided new capabilities of dynamic provisioning, monitoring and managing resources by providing immediate access to resources, enabling easy scaling up of services and implementation of new classes of existing applications. However, sharing the same pool when requesting services involves the risk of data breaches, account compromises, injection vulnerabilities, abusive use of features such as the use of trial periods and distributed denial of service (DDoS) attacks. 3,4 As a result, many customers rank cloud security as a major challenge that threatens their work and reduces their trust in cloud service providers. Cloud-based architectures have reduced IT barriers and provided new capabilities of dynamic provisioning, monitoring and managing resources by providing immediate access to resources, enabling the easy scaling up of services. However, sharing the same pool when requesting services involves the risk of data breaches, account compromises, injection vulnerabilities and distributed denial of service (DDoS) attacks. As a result, many customers rank cloud security as a major challenge that threatens their work and reduces their trust in cloud service providers. Amar Meryem and Bouabid EL Ouahidi propose an architecture that eradicates malicious behaviours by detecting known attacks using log files; blocks suspicious behaviours in real time; secures sensitive data; and establishes better adaptations of security measures by dynamically updating security rules.