利用关联规则作为恒星光谱数据相关性分析方法,以VC++和Oracle9i作为开发工具,设计并以程序实现了基于关联规则的恒星光谱数据相关性分析系统,给出了其软件体系结构和模块功能,并对恒星光谱数据的预处理、取样技术、背景知识模板、频繁模式集挖掘算法等关键技术进行了详细描述。最后,通过系统运行结果表明,利用关联规则来描述、分析光谱数据特征和物理化学性质之间存在的相关性是可行的和有价值的。
In this paper, the interrelationship analysis system on star spectrum data is designed and implemented by using the association rule as the way of analyzing star spectrum data and VC ++ and Oracle9i as the development tools, and the involved software architecture and functional modules are elaborated. In particular, the following key techniques are discussed in details: preprocessing of star spectrum data; sampling techniques; background knowledge templates; and association rules mining. In the end, the system's experimental results show that adopting association rule to discuss the above interrelationships is feasible and valuable.