基于Markovgame模型的网络故障序列模式挖掘算法  

Mining algorithm of network fault sequence pattern based on markovgame model

作　　者：李俊磊[1] LI Junlei(Guangdong Justice Police Vocational Collegey Guangzhou 510520,China)

机构地区：[1]广东司法警官职业学院,广州510520

出　　处：《自动化与仪器仪表》2020年第9期170-172,177,共4页Automation & Instrumentation

基　　金：广东司法警官职业学院项目“基于大数据环境下数据挖掘在电子证据中的研究与应用”(No.QN20170001)。

摘　　要：针对传统的网络故障序列模式挖掘算法无法挖掘多层次网络的问题,提出一种基于Markovgame模型的网络故障序列模式挖掘算法。首先使用Markovgame模型挖掘故障序列数据,计算模型定阶ARMA(p,q)中的AIC信息准则,依据计算出的结果得到故障序列模式数据值,预处理得到的数据值,将得到的故障数据值经清洗、变换、集成、规约4部分预处理后,确定挖掘算法参数,不断调整预处理得到的数据特定区间,利用故障数据特定区间中的学习因子,得到挖掘算法中的参数,计算参数之间的相关系数,得到多维故障序列模式数据规则集后,实现故障序列模式的挖掘。实验结果表明:与传统网络故障序列模式挖掘算法相比,基于Markovgame模型的网络故障序列模式挖掘算法可以挖掘到4层的网络故障序列模式,比传统网络故障序列模式挖掘算法挖掘到的网络层更多,适用性更好。Aiming at the problem that traditional network fault sequence pattern mining algorithm can’t mine multi-level network,a network fault sequence pattern mining algorithm based on markovgame model is proposed.Firstly,we use markovgame model to mine fault sequence data,calculate AIC information criterion in ARMA(P,q)of model order,get fault sequence pattern data value according to the calculated result,pre-processing the data value,after cleaning,changing,integrating and pre-processing the fault data value,determine the mining algorithm parameters,and adjust the pre-processing data continuously According to the specific interval,using the learning factors in the specific interval of the fault data,the parameters in the mining algorithm are obtained,the correlation coefficient between the parameters is calculated,the multi-dimensional fault sequence pattern data rule set is obtained,and then the mining of fault sequence pattern is realized.The experimental results show that:compared with the traditional network fault sequence pattern mining algorithm,the network fault sequence pattern mining algorithm based on markovgame model can mine four layers of network fault sequence pattern,which is more applicable than the traditional network fault sequence pattern mining algorithm.

关 键 词：Markovgame模型 网络故障 序列模式 挖掘算法 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]