Combined Forecasting Model of Network Security Situation Based on Information Fusion
Yong Kang Peng, Jianhuan Huang, Zhe Wang · 2019
In this paper, to control the development direction of network security situation more accurately and solve the forecasting problems in network security situation awareness (NSSA) a combined forecasting model of network security situation based on information fusion was proposed, in which sample set was firstly built according to the forecasted value and actual value of the situation of single forecasting model, and then the data set was used to calculate weight index, evaluation index and weight distribution of single forecasting model, and next the fusion characteristic of evidence theory was used to get the combination weight and carry out the combined forecasting, and finally each index was optimized by evidence discount method. The simulation results showed that the model can forecast the network security situation more effectively with the accuracy improved by about 8% on average compared with the existing forecast algorithms.