本文選題：網(wǎng)絡(luò)安全態(tài)勢預(yù)測 + CMA-ES優(yōu)化算法　； 參考：《哈爾濱理工大學(xué)學(xué)報(bào)》2017年02期

【摘要】：針對網(wǎng)絡(luò)安全態(tài)勢預(yù)測問題,提出了一種預(yù)測方法。該方法采用協(xié)方差矩陣自適應(yīng)進(jìn)化策略(CMA-ES)算法來優(yōu)化徑向基神經(jīng)網(wǎng)絡(luò)(RBF)預(yù)測模型中的參數(shù),使得RBF預(yù)測模型具備更好的泛化能力,可以快速的找出復(fù)雜時(shí)間序列中的規(guī)律。仿真實(shí)驗(yàn)結(jié)果表明,采用CMA-ES優(yōu)化的RBF預(yù)測模型能夠準(zhǔn)確預(yù)測出一段時(shí)間內(nèi)的網(wǎng)絡(luò)安全態(tài)勢值,預(yù)測精度高于傳統(tǒng)預(yù)測手段。
[Abstract]:Aiming at the problem of network security situation prediction, a prediction method is proposed. In this method, the covariance matrix adaptive evolutionary strategy (CMA-ESS) algorithm is used to optimize the parameters in the prediction model of radial basis function neural network (RBFN), so that the RBF prediction model has better generalization ability and can quickly find out the rules in the complex time series. The simulation results show that the RBF prediction model optimized by CMA-ES can accurately predict the network security situation value for a period of time, and the prediction accuracy is higher than that of the traditional prediction method.
【作者單位】： 長春工業(yè)大學(xué)應(yīng)用技術(shù)學(xué)院;海南師范大學(xué)信息科學(xué)技術(shù)學(xué)院;長春市十一高中信息技術(shù)教研組;
【基金】：吉林省教育廳科學(xué)技術(shù)研究項(xiàng)目(吉教科合字[2014]第145號,[2016]第344號) 海南省自然科學(xué)基金面上項(xiàng)目(617120,617121)
【分類號】：TP393.08
，

本文編號：1966278