Research on Intrusion Detection Model Based on Bagged Tree

Abstract

Intrusion detection is an important step to ensure network security. Under today’s network environment, because of the increasing complexity of network security issues and the increasing amount of network security data, intrusion detection systems require higher recognition accuracy. The accuracy and effectiveness of traditional methods of intrusion detection no longer meet the need of today’s big data era. In the past ten years, the performance of the classifier is further improved with the development of machine learning, so we chose to apply machine learning on intrusion detection. In this paper, we propose a new intrusion detection method, which is based on Bagged Tree, and the data set UNSW_NB15 is used to verify the model. Experimental verification proves that the model designed in this paper has higher detection accuracy than previous classic intrusion detection algorithms.