在线网络异构容错数据的同构漏洞挖掘仿真

采用当前方法挖掘异构容错数据中存在的同构漏洞时,不能有效的去除网络数据中存在的噪声,挖掘同构漏洞所用的时间较长,存在去噪效果差和挖掘效率低的问题。提出在线网络异构容错数据的同构漏洞挖掘方法,在经验模态分解方法的基础上采用集成经验模式分解方法对在线网络数据做去噪处理,利用差分法抑制在线网络数据中存在的粗差干扰,抑制并分解网络中存在的脉冲干扰,分层去除数据中存在的噪声。提取去噪处理后的数据集中的元组,并对元组作概化处理,获得高层属性,根据高层属性划分网络数据,将同构数据划分到一起,实现在线网络异构容错数据中同构漏洞的挖掘。仿真结果表明,所提方法的去噪效果好,挖掘效率高。

Homogeneous Vulnerability Mining Simulation of Online Network Heterogeneous Fault Tolerant Data

At present, when the method is used to mine isomorphic vulnerabilities in heterogeneous fault-tolerant data, the noise in network data cannot be effectively removed. Therefore, an isomorphic vulnerability mining method for heterogeneous fault-tolerant data in online network was proposed. On the basis of empirical mode decomposition, an integrated empirical mode decomposition method was used to remove the noise of online network data. The difference method was used to suppress gross errors in online network data and decompose impulse interference in network, so as to remove noise in data hierarchically. Moreover, the tuples in denoised data set were extracted, and the tuples were generalized to obtain the high-level attributes, which were used to represent the clustering results. The network data were divided by the high-level attributes, and the isomorphic data were divided into the same class. Finally, the mining of isomorphic vulnerabilities in heterogeneous fault-tolerant data of online network was realized. Simulation results show that the proposed method has good denoising effect and high mining efficiency.