Abstract: The rule sets extracted by traditional algorithm are usually very large,because a number of rules can be generated by other rules.The number of rules can be reduced using closed item sets.The relationship of generalization and specialization among concepts of concept lattice is very suitable for extracting rules.Now several kinds of algorithms for extracting rules based on concept lattice centered on getting non-redundant rules that have accurate support and confidence.The algorithm that extract rule-generating set based on concept lattice with which people can generate all frequent and confident rules can reduce number of rules and is more efficient.This paper introduces a kind of data structure that is used storing the rules and the algorithm that can lead to the general rule-generating set from rule-generating set.

Key words: rule-generating set, concept lattice, association rules, rule generating, rule extracting

摘要： 数据挖掘中传统的关联规则生成算法产生的关联规则集合相当庞大,其中很多规则可由其它规则导出。使用闭项集可以减少规则的数目,而概念格节点间的泛化和例化关系非常适用于规则的提取。目前几种基于概念格的规则提取算法局限于得到准确支持度、信任度的无冗余规则。提出了一种在概念格上挖掘出能推导出所有满足最小支持度、信任度规则的规则产生集算法,文中称之为组规则产生集算法,减少了规则的规模。在此基础上进一步给出了组规则产生集的存储数据结构并用其导出一般规则产生集的算法。

关键词: 规则产生集, 概念格, 关联规则, 规则推导, 规则提取