METHOD AND APPARATUS FOR DETECTING ABNORMAL BEHAVIOR BY ANALYZING BEHAVIOR PATTERN OF USER IN VARIOUS WEB SERVICE ENVIRONMENTS

摘要

The present invention relates to a method for detecting abnormal behavior and, more specifically, to a method and an apparatus for detecting abnormal behavior through profile-based analysis and machine learning-based analysis, adding or removing an abnormal behavior analysis rule by using a behavior parameter value of a user, and generating an abnormal behavior analysis model by machine learning to detect abnormal behavior. According to an embodiment of the present invention, the method for detecting abnormal behavior comprises: a step of collecting a behavior parameter value of a user in real time; a first detection step of comparing the collected behavior parameter value of the user with an abnormal behavior analysis rule of the corresponding user to detect abnormal behavior; a second detection step of inputting a behavior parameter value of the user which is not determined as abnormal behavior in the first detection step into an abnormal behavior analysis model to detect abnormal behavior by calculation of the abnormal behavior analysis model; and a step of combining a detection result of the first detection step and a detection result of the second detection step to determine abnormal behavior.;COPYRIGHT KIPO 2017