基于变量分组贝叶斯网络的安全态势评估方法

A security situation assessment method based on subdividing Bayesian network in sub-networks

针对网络安全威胁问题,将人工智能理论和相关技术与网络安全态势评估相融合,提出一种以细化变量进行分组的贝叶斯网络作为基础研究的网络安全态势评估方法。该算法可以有效减少变量数量,缩短产生贝叶斯网络的程序运行时间,并通过相关实验验证了有效地减少变量数量对最终的结果并没有产生过多影响。用本算法对大量网络实际运行数据进行测试,结果表明该方法能够很好地区分不同的网络安全威胁,从而能够有效评估网络安全态势。

Aiming at the problem of security situation awareness about networks, using artificial intelligence theory and related technologies com- bined with network security situation assessment, a network security situation awareness method based on subdividing Bayesian network is proposed. The method can effectively reduce the number of variables, shorten the running time in the progress of generating Bayesian networks. The experiment proves that the method effectively reduces the number of variables but it dosen＇ t have too much influence on the final results. Using the proposed algorithm, a large number of the networking operation data were tested, the experiments results show that the method is able to distinguish different network security threats so as to effectively evaluate the network security situation.