摘要
给出了时间序列的变化序列和最近时间子序列的概念.并把最近时间子序列看作是时间序列的信息聚集器。基于这一现点,提出了基于变化序列的时序数据挖掘技术。先把待挖掘的时间序列转换成时间序列的变化序列,然后利用时间序列的变化序列的最近时间子序列隐含的知识,以指导对原时间序列的挖掘,并给出了挖掘算法。最后算例验证了该挖掘方法的可行性和有效性。
The time series varying series and the latest time sub-series are defined. The latest time subseries is regarded as the information collector of the time series. Based on the concept,the time series data mining technique based on time series varying series is presented. Firstly the time series set for mining is converted into its varying series set, and then the information underlying in the latest time sub-series of the time series varying series is used as a guide to excavate the original time series. And mining algorithms are given. Finally,a concrete mining example proves that the method is feasible and effective.
出处
《南京航空航天大学学报》
EI
CAS
CSCD
北大核心
2006年第B07期154-157,共4页
Journal of Nanjing University of Aeronautics & Astronautics
基金
国家自然科学基金(60461001
60574052)资助项目
广西教育厅科研基金(200547)资助项目。
关键词
时序数据挖掘
变化序列
最近时间子序列
time series data mining
varying series
latest time sub-series
作者简介
王勇,男,博士,副教授,1963年生,E—mail:wangyong@gxun.cn。