【作者】 郭辉辉；

【导师】 崔广才；

【作者基本信息】 长春理工大学 ， 计算机应用技术， 2011， 硕士

【摘要】 近年来,随着网络技术的发展,数据量的飞速增长与信息量的日益缺乏两者之间相互矛盾,数据挖掘技术越来越被人们所关注。纵观各种数据挖掘技术,关联规则挖掘已经成为数据挖掘方向的一个重要研究课题。目前最常用的就是利用关联规则算法来挖掘Web日志来发现用户的访问规律,兴趣爱好,从而实现对不同的用户提供不同的服务,达到个性化服务。本文首先简要介绍了数据挖掘技术的产生、Web数据挖掘的基本概念、分类、挖掘过程及常用的几种相关技术,详细讨论了数据预处理的过程；其次阐述了关联规则.的概念,详细了分析了关联规则的经典算法——Apriori算法；再次详细讨论了关联规则的增量数据挖掘算法,并且在此基础上,分析和总结其算法的不足之处,本文提出了基于划分的FUP算法,提高关联规则挖掘的效率；最后通过实验实现了基于划分的FUP算法,并且将实验结果与传统的FUP算法进行比较分析,从而验证了基于划分的FUP算法的有效性和正确性。更多还原

【Abstract】 In recent years, with the development of network technology, the rapid growth data and the increasing lack of information conflicts with each other and data mining has been growing concerned. Throughout the various data mining techniques, association rule mining has become an important research topic of the direction of data mining. Today we usuallly mine Web logs to discover the access rules and hobbies of users by the use of association rule mining algorithms in order to personalize service that the different users have different services.First, this article briefly describes the generation of data mining, the basic concepts, classification, data mining process of Web data mining and several related technologies that are largely used and discusses in detail the process of data preprocessing; Second, this article describes the concept of association rules, analysis in detail the classical algorithm of association rules—Apriori algorithm;Also the paper discusses in detail the incremental data mining algorithm of association rules and analysis and summary the inadequacies of the algorithm on the base. This paper proposed FUP algorithms based on classification to improve the efficiency of association rule mining; Last the FUP algorithms based on classification is come true by experiments and the validity and accuracy is verified when experimental results were compared with the traditional FUP algorithm.更多还原