一种改进的最大频繁集发现算法

关联规则是数据挖掘的主要技术，而最大频繁集是关联规则挖掘的核心。关联规则发现的准确性与效率的好坏直接决定了发现的知识规则是否适用。阐述了关联规则、频繁集和频繁超集的定义，分析了现有关联规则算法的思想及其不足，然后在概率的基础上引入了期望长，提出了ELMFI算法，最后用实例进行仿真实验并做了比较分析。该算法直接产生期望长度的候选项集并进行验算，试验结果验证了其可行性，发现效率有所提高，能节约大量的系统空间和运算时间。

An Improved Algorithm of Maximum Frequent Itemsets

Association rules is the main technique for data mining,maximum frequent itemset is the key of association rules.The accuracy of the association rules and the quality of the efficiency come to a decision of whether knowledge rule apply or not.First elaborate the definition frequent itemset,frequent super itemset and association rules,analyze the thought and its shortage of the association rules.Then import expectation length based on probablity,put forward ELMFI algorithm.Finally,carry on analysis by using solid examples.This algorithm could economize a great deal of system space and the operation time,suitable for the large database.