ARTIFICIAL INTELLIGENCE-BASED SECURITY EVENT ANALYSIS SYSTEM THROUGH SEMI-SUPERVISED LEARNING AND METHOD THEREOF

摘要

The present invention relates to an artificial intelligence-based security event analysis system through semi-supervised learning and a method thereof. More specifically, in an artificial intelligence-based security event analysis system though machine learning, the security event analysis system through semi-supervised learning minimizes a time to label classified security events by sampling unlabeled data and conducting the semi-supervised learning on problems in which learning data required for supervised learning is insufficient, and increases accuracy of a security policy to, as a result, increase efficiency of security control and minimize over-detection or false detection in the security events.;COPYRIGHT KIPO 2020