基于PCA-BP的信息安全风险评估模型

Model of Information Security Risk Assessment Based on PCA-BP

在网络安全问题的研究中,由于黑客通过操作系统进行攻击,造成网络信息不安全。为了提高信息安全风险评估的准确性,提出了基于主成分分析(PCA)和BP神经网络的信息安全风险评估模型。首先运用层次分析法对影响信息安全风险因素进行定性和定量的分析;然后采用主成分分析对影响信息安全风险全部因素重新组合生成新的综合指标;最后采用BP神经网络建立信息安全风险评估模型,并使用Matlab进行仿真,预测其风险值。仿真结果表明基于PCA-BP的评估模型准确率(98.48%)比基于BP评估模型(96.41%)的高些。

In order to improve the accuracy of information security risk assessment （ISRA）, an ISRA model based on PCA - BP was proposed. First, all factors affecting information security risk assessment were analyzed by using analytic hierarchy process （AHP） that combines the ways of qualitative methods and quantitative methods. Second, the factors were recombined to generate new comprehensive indexes by principal component analysis （PCA）. Third, the information security risk assessment model based on PCA - BP was established to predicate risk values by BP. The experimental results indicate that the accurate rate （98.48%） of the assessment model based on PCA - BP is higher than that （96. 41% ） of the assessment model based on BP.