组合核函数相关向量机的网络安全态势预测

为了提高网络安全态势的预测精度，针对单一核函数的局限性，提出一种组合核函数相关向量机的网络安全态势预测模型。首先对网络安全态势时间序列进行重新构造，得到相关向量机的学习样本，然后采用多项式和高斯核函数构建组合核函数，并采用组合核函数相关向量机对网络安全态势样本进行学习，建立网络安全态势预测模型，最后对网络安全态势预测性能进行测试。实验结果表明，相对于单一核函数相关向量机以及其它网络安全态势预测模型，组合核函数相关向量机提高了网络安全态势的预测准确性，可以满足网络安全态势预测的实际应用需求

Network security situation forecasting based on combining kernels function relevance vector machine

In order to improve the prediction accuracy of network security situation, due to limitations of single kernel function, this paper puts forward a network security situation forecasting model based on combining kernels function relevance vector machine. Firstly, the network security time series are structure to get learning samples of the relevant vector machine, and then the combined kernel function is constructed with polynomial and Gauss kernel function, and relevance vector machine with combined kernel function is used to establish the forecast model of network security situation, finally the performance of the model is tested. The results show that, compared with the single kernel relevance vector machine and other network security situation forecasting models, relevance vector machine with combined kernel function has improved the prediction accuracy of network security situation, and can be used to meet the prediction needs of network security situation.