摘要
模糊集与本体结合的数据挖掘方法得到了广泛的关注。为了丰富数据挖掘效果以及数据挖掘得出的规则的完整性,本文在模糊本体的挖掘算法基础上,提出了模糊本体中叶子结点的相似度定义以及不同语义层次所含项目集的数目定义多重最小支持度,提出了基于模糊本体的广义关联规则算法。对比实验证明,基于模糊本体的广义关联规则算法的挖掘具有更强的可读性,获得的语义关联规则更加丰富,促进了在广义关联规则挖掘过程中使概念泛化更加合理,提高了算法效率。
In data mining,the method of fuzzy set based on fuzzy ontology is hot topic. The method based on the improved fuzzy ontology with the definition similarity of the leaf node, and the algorithm of generalized items based on multiminsupport frequency enrich the effect and integrity of data mining. It is proved by the experiment that the generalized concepts are more reasonable, and acquired semantic association rules are more abundance.
出处
《计算机工程与科学》
CSCD
北大核心
2009年第9期105-107,121,共4页
Computer Engineering & Science
基金
安徽省自然科学基金资助项目(050420204)
安徽省高校自然科学研究基金资助项目(2006kj055B)
关键词
数据挖掘
模糊本体
关联规则
多重最小支持度
data mining
fuzzy ontology
association rule
multiple minimum supports
