针对目前数据挖掘系统缺乏通用性和复用性的问题,对UML进行轻量级扩展,采用UML Profile机制建立了一套关联规则挖掘元模型,实现了关联规则挖掘在概念层上的建模设计,取代了以往在具体的表结构和数据仓库系统上进行建模的方法,并在某大型钢结构企业的决策支持系统中验证了模型的有效性。最后在Analysis Services 2008上经过验证,利用UML Profile机制建立的关联规则挖掘元模型可较早地伴随决策系统进入设计阶段,减少开发的时间和代价。
Aiming at lack generality and reusability in data mining system, established a set of association rule mining metamodel by using the lightweight extension mechanism of UML. With this profile, replaced the traditional method which considering on the final table structure and DWs. And finally put it into decision making support system of a large steel company. Practice proves that the metamodel can be easily used in the design of the decision making system,thus reduces the time and cost.