Security System for Cloud Service Based on Machine Learning

摘要

The present invention relates to a machine learning-based cloud service security system, comprising: a multi-cloud adapter that interworks with a plurality of cloud services to receive the cloud service-related information; an analysis unit configured to collect the cloud service-related information through an API call to the multi-cloud adapter and detect malicious actions included in the information; and a management unit for monitoring an analysis result of the analysis unit or controlling analysis settings of the analysis unit; Including, wherein the analysis unit is characterized in that the malicious behavior detection is performed by fusing the pre-stored security settings for detection of malicious actions and the security settings for detection of malicious actions predicted based on machine learning. Accordingly, information provided from a plurality of cloud services can be integratedly collected through a multi-cloud adapter and applied to security settings derived based on machine learning to maximize the accuracy of malicious behavior detection.