Traditional security situational awareness methods do not consider attack probability,resulting in one-sided percep-tion results.Therefore,an improved network security situational awareness method based on RBF neural network is proposed.Uti-lize web crawlers to collect network data and perform multi-source data fusion processing.The fused data is normalized and used as input for the improved RBF neural network to calculate the attack probability.Analyze network traffic data,extract features and re-duce dimensionality,and screen the most influential feature set.Combining machine learning models for node clustering analysis,constructing a security situation rating function,and calculating network security situation values.Comparative experiments show that this method can accurately perceive the network security situation.
万方数据