一类相邻时态关联规则的知识发现问题  

KNOWLEDGE DISCOVERY FOR A CLASS OF CONTIGUOUSLY TEMPORAL ASSOCIATION RULES

作　　者：孟志青[1] 

机构地区：[1]湖南湘潭大学信息工程学院计算机系,湘潭411105

出　　处：《模式识别与人工智能》2001年第4期458-462,共5页Pattern Recognition and Artificial Intelligence

基　　金：湖南省教育厅科研资助项目

摘　　要：本文研究了一类新的知识发现问题即基于时态约束的相邻时态关联问题,首先我们引进了时间的时态型、基时态型等概念及其性质,然后我们给出了有限个属性在时态型的基础上描绘不同状态的时态事件空间,从而定义了事件之间的时态关联规则,由此导出了1种具有一定意义的相邻时态关联规则,这种相邻时态关联规则可以用于商品销售、股票价格等问题的知识发现和短期的预决策,最后讨论了在2个时态型下的相邻时态关联规则的算法主要步骤,并给出了实例的分析。In this paper we discuss a kind of contiguously temporal association rules based on temporal constraint. First, we give an introduction to concepts of temporal type and basis temporal type. Then, we define a temporal event space describing various states of events based on different attributes of the temporal type and the temporal association rule between events, which imply one class of contiguously temporal association rules with special significance. This can be applied to many problems of knowledge discovery and prediction and decision in the shorter time, such as business sale and stock pricing. Finally, we give the main idea of our algorithm for the contiguously temporal association rule with two temporal types and analysis it with an example.

关 键 词：知识发现 数据挖掘 时态型 相邻时态关联规则 人工智能 数据库 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论] TP18[自动化与计算机技术—计算机科学与技术]