Cloud Data Center Intrusion Detection Model Based on Active Rules
Wei Zhao, Xiaoming Jiang, Jingchun Wang · 2020
Because the part of rules matching takes up a relatively high proportion in the current intrusion detection model and the rules adjustment can also influence the data accuracy, this paper proposes an anomalous detection model based on the active rules. Aiming at the problem of low rules adjustment efficiency in the current model, this paper designs the structure of active rules and a dynamic adjustment approach of active rules based on two-steps. This paper selects rules matching approach to update the matching process dynamically on the basis of activeness, and thus reducing the time complexity of intrusion detection system and false alarm rate. The experimental results indicate that the anomalous detection model relying on active rules proposed here can further improve the efficiency of rules matching and reduce the false alarm rate, performing a stronger practicability.