Machine Learning-Based Research on Network Security\'s Primary Technologies

Abstract

Abstract: A large number of network devices, applications, and the explosive growth of network data have made the network environment extremely complex with significant potential hazards to network security. This is due to the ongoing development of information technology. Cyber attackers are now attacking network settings with connected histories rather than just common users; examples of these environments include businesses, governments, and nations. Massive amounts of Internet data have been produced by the diversification of network services, and traditional network security systems have had trouble keeping up with the demands of network security in terms of performance and self-adaptability. The growth of concepts in the field of network security has been greatly aided by research on machine learning-based network security that has produced numerous findings, demonstrating strong skills in processing large data, automatic learning, detection, and identification. In this paper, we integrate machine learning-related technologies to enhance the performance of intrusion detection and alarm correlation automation, and we investigate key technologies such as machine learning-based network security situational awareness methods and dynamic data stream classification methods based on judgment feedback, to enhance the detection performance, adaptive and generalization capabilities of machine learning-based network security