基于支持向量机的Web日志频繁序列模式挖掘研究

Research on Web log frequent sequence pattern mining based on SVM

为了降低Web日志频繁序列模式挖掘误差,提出基于支持向量机的Web日志频繁序列模式挖掘方法。构建Web日志频繁序列模式检测序列,采用自相关特征分布式融合方法进行序列重组,提取序列模式的统计特征量,对其特征分布值进行信息融合。建立Web日志频繁序列模式融合式调度模型,采用支持向量机分析方法进行Web日志频繁序列模式挖掘的自适应学习与寻优控制,实现Web日志频繁序列模式挖掘。仿真结果表明,采用该方法进行Web日志频繁序列模式挖掘的误差较低,收敛性较好。

In order to reduce the error of mining frequent sequential patterns of Web logs,a method of mining frequent sequential patterns of Web logs based on SVM is proposed.This paper constructs the frequent sequence pattern detection sequence of Web logs,recombines the sequence by using the distributed fusion method of autocorrelation features,extracts the statistical features of sequence patterns,and fuses the feature distribution values.A scheduling model of Web log frequent sequence pattern fusion is established.The adaptive learning and optimization control of Web log frequent sequence pattern mining are carried out by using SVM analysis method.Simulation results show that this method is used to mine frequent sequence patterns of Web logs with low error and good convergence.