【摘要】 在基于隐马尔可夫模型的网络安全实时风险评估中,状态转移概率矩阵的确定是关键一步,目前基本上都是依据经验给出,具有很大主观性,不能客观反映网络安全的风险状况。为此,引入了攻击难度系数的概念,通过对数据集的统计学习,给出了状态转移概率矩阵。此外,通过对威胁进行分类,根据各类威胁的影响,给出了相应的权重。实验结果表明,该方法使得网络安全实时风险评估更加客观,为网络安全的风险管理提供了决策支持。更多还原

【Abstract】 In the field of network risk assessment based on hidden Markov models,the state transition matrix are usually derived from expert experience.It leads to much subjectives to the result of the assessment which can not objectively reflect the real risk of the network. Therefore,a conception of attack difficulty coefficient is introduced.Through analyzing and statistical learning the dataset,the state transition matrix is derived.Taxonomy of those threats are done in the dataset,then weights all of them according to their influence. The experiment indicates that the result of the assessment using the method proposed is much more objective while providing better support for network risk management.更多还原