摘要
数据挖掘是从大量的数据中提取知识。数据挖掘的主要功能有关联、分类、预测、聚类和时序分析等。文章对关联规则和分类规则这两种挖掘模式的存储方法进行研究,提出在关系数据库系统中使用主-子表来存储关联规则和分类规则的方法,继而将这种存储方法应用在其它类型的模式上,形成一种统一的数据挖掘模式的存储方法。
Data mining is a process to refine the information and knowledge from large amount of data. The main functions of data mining include association, classification, clustering, prediction and time-series analysis etc. This paper mainly discusses the storage methods of two major mining patterns-association rules and classification rules. Then it refines the method of using main-table and sub-table to store association rules and classification rules in a relational database system. Afterwards we can apply it to other relative patterns, forming a new unified storage method of data mining patterns.
出处
《昆明大学学报》
2006年第4期44-47,共4页
Journal of Kunming University
关键词
数据挖掘
存储方法
模式
关联规则
分类规则
主-子表
data mining
storage method
patterns
association rule
classification rules
main-table and sub-table
