从Fuzzy Taxonomic数值型数据库中挖掘一般化关联规则

挖掘关联规则是数据挖掘研究的一个重要方面．基于属性内通常还存在更高层次的抽象，即呈现出Taxonomic结构这一事实，Srikant和Agrawal等人提出了在确定的Taxonomic结构下挖掘泛化布尔型关联规则的挖掘算法．但在实际应用中，往往这种Taxonomic结构还呈现出模糊性；着重研究了在这种模糊Taxonomic结构下如何从数值型数据库中挖掘一般化关联规则的问题，提出了，一种新的Fuzzy Taxonomic数值型数据库模型，并提出了相应的规则发现方法，两个实例数据库表明了新模型的有效性和灵活性．

ining Generalized Association Rules from Fuzzy Taxonomic Quantitative Databases

Mining association rules is a major aspect of data mining research In practice, there are multiple levels of abstraction (i e, taxonomic structure) among the attributes of the databases Srikant and Agrawal have proposed several algorithms to mine generalized Boolean association rules upon all levels of presumed crisp taxonomic structures However, in many real world applications, the taxonomic structures may not be exact but fuzzy This paper focuses on the issue of how to mine generalized association rules from quantitative database with fuzzy taxonomic structure, and a new fuzzy taxonomic quantitative database model is proposed to solve the problem Finally, the simulations verify the effectiveness of the new model