基于模糊聚类的分布式Web日志挖掘方法  

Distributed Web Log Mining Method Based on Fuzzy Clustering

作　　者：陈宝国 宋旸 CHEN Baoguo;SONG Yang(School of Computer Science,Huainan Normal University,Huainan 232000,China)

机构地区：[1]淮南师范学院计算机学院,安徽淮南232000

出　　处：《太原师范学院学报（自然科学版）》2020年第3期54-58,共5页Journal of Taiyuan Normal University:Natural Science Edition

基　　金：2018年安徽高校自然科学重点研究项目(KJ2018A0469);淮南师范学院2019年度校级科学研究项目(2019XJYB14)。

摘　　要：为了提高分布式Web日志挖掘能力,提出基于模糊聚类的分布式Web日志挖掘方法.构建分布式Web日志的关联规则分布集,采用模糊信息聚类分析方法进行分布式Web日志关联规则特征聚类处理,提取分布式Web日志性的多重关联特征量,结合重叠性迭代检测方法进行分布式Web日志挖掘过程中的自适应寻优,采用模糊关联规则调度方法进行分布式Web日志挖掘的负载均衡调度.通过计算邻接点的适应度函数,对相似度高的分布式Web日志关联规则进行合并处理,根据模糊信息聚类结果实现分布式Web日志挖掘优化.仿真结果表明,采用该方法进行分布式Web日志挖掘的精度较高,提高了分布式Web日志的推荐和信息检索能力.In order to improve the ability of distributed Web log mining,a distributed Web log mining method based on association rules and fuzzy clustering is proposed.Build distributed Web log distribution set of association rules,fuzzy clustering analysis method for distributed Web information log feature clustering,association rules extraction distributed Web log multiple correlation characteristics,combined with overlaps iterative detection method for distributed adaptive optimization of the process of Web log mining,scheduling method using fuzzy association rules on distributed load balance scheduling of the Web log mining.By calculating the fitness function of the adjacency points,the distributed Web log association rules with high similarity are combined,and the distributed Web log mining optimization is realized according to the fuzzy information clustering results.Simulation results show that the precision of distributed Web log mining using this method is high,and the ability of distributed Web log recommendation and information retrieval is improved.

关 键 词：关联规则 模糊聚类 分布式 WEB日志 挖掘 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]