一种无冗余的Web日志挖掘算法

Web日志挖掘是Web数据挖掘的一个重要研究领域。Web日志挖掘通过发现Web日志中用户的访问规律和模式,可以提取出其中潜在的规律和信息,人们对这个领域的研究也日益重视。然而,传统的基于关联规则的Web日志挖掘算法都是基于所有关联规则的。这种方式往往挖掘产生大量的候选规则,而且存在大量冗余的规则。提出了一种新的无冗余的Web日志挖掘算法,该算法通过引入频繁闭项集合最小关联规则的概念,从而解决了以往基于所有关联规则挖掘算法中出现的上述问题。

A New Web Log Mining Algorithm without Redundancy

Web log mining is one of the important research areas in Web data mining,through which the access law and mode of the user could be found. More attention is also paid to the Web log mining. However, the traditional Web log mining algorithm are all based on all of the association rules. And the traditional way often produces not only large amounts of the candidate frequent itemsets, but also lots of redundant rules. This paper puts forward a new Web log mining algorithm without redundancy,and the concept of frequent closed itemsets and minimal association rules are proposed to solve the problems appeared in mining Web log based on all frequent itemset association rules.