Network Security Maintenance and Detection Based on Diversified Features and Knowledge Graph

Due to the large processing capacity and high false alarm rate, current security defense technologies in enterprise networks are often deployed in practical environments, resulting in weak defense capabilities. Based on this, a proactive defense architecture is proposed based on diverse features and knowledge graphs, and its performance is analyzed. The experimental results showed that the detection rate of the detection method in malicious attack behavior could exceed 85% or more. The false alarm rate was significantly less than 10%. The detection rate of 9 types of malicious attack software remained between 76.1% and 98.7%. The detection method in normal access behavior detection had a minimum false alarm rate of 0.05% and a maximum false alarm rate of only 0.1%. The proactive defense effect of the proxy server was relatively obvious, which was more effective in blocking traffic data. At the same time, the defense effectiveness of proactive defense architecture varied under different attack methods, reaching a maximum of 98%, which was basically consistent with the detection rate in malicious attack behavior. Overall, the proactive defense architecture for network security maintenance and detection is effective and practical, which can be effectively applied in practical enterprise network security maintenance and detection.