SYSTEM AND METHOD FOR DETECTING ABNORMAL BEHAVIOR BASED ON UNIFIED MODEL

摘要

The present invention relates to a system for sensing an abnormal behavior based on a unified analysis model and a method thereof. An embodiment of the present invention provides a system for sensing abnormal behavior, comprising: an information receiving unit receiving usage information in accordance with usage of digital content; a model generation unit generating an analysis model for machine-learning the received usage information to sense the abnormal behavior in accordance with usage of the digital content; a model transmission unit transmitting the analysis model to a cloud server; and an abnormal behavior sensing unit receiving a unified analysis model from the cloud server, extracting a portion from the unified analysis model based on a business policy or a customer policy of each company, generating an analysis model assembled by unifying the extracted partial analysis model with the generated analysis model and sensing the abnormal behavior based on the assembled analysis model, wherein the cloud server collects the transmitted analysis model and machine-learns the collected analysis model to generate the unified analysis model. The present invention makes each analysis server transmit only the analysis model, acquired through machine learning, to the cloud server, thereby minimizing traffic in accordance with data transmission.