A Security Situation Awareness System based on Wide & Deep

Abstract

Security situation awareness analysis is trending to be an important part of the cyber security currently. To assess the network security situation, a security situation awareness system based on the distributed cluster model is proposed in this paper. The distributed cluster model of network security situation awareness is built based on the improved wide & deep model of TensorFlow. The PRelu activation function is introduced to make the model more fitting and the weight attenuation of the Softmax loss function is added to improve the accuracy effectively and reduce the time of the situation prediction process. In addition, we implement the prototype and evaluate the effectiveness and usability of the situation awareness system with the database provided by a safety monitoring system applied in State Grid Corporation of China. The experimental results demonstrate that the system we proposed in this paper improves the accuracy of the situation prediction effectively and reduces the time of the situation prediction.