基于时空快照数据库的时间序列预测

时空数据库是在空间数据库的基础上引入了时间维,时空数据模型和时空变化分析是GIS领域当前研究热点之一。提出一种在时空快照数据中预测时间序列发展和关联规则发现的方法。首先采用基态修正模型表达时空数据,从中提取出时空快照序列,将时空快照序列聚类为几个簇,再在簇内挖掘关联规则。将该方法应用于实验数据,结果证明这种方法能够有效地从时空快照数据中发现时空序列的发展趋势。

Based on space-time snapshot of database time series prediction

Spatio-temporal databases is the database on the basis of the introduction of a time dimension to space database,spatio-temporal data model and the analysis of time and space changes in the field of GIS is the current hot research topic.This paper presents a snapshot of data in space and time predicted time series development and association rules approach.First use of the ground state that the expression spatio-temporal data models,extracted from the space-time snapshot sequence,snapshot sequence of space-time clustering of several clusters,and clusters within the mining association rules.The method is applied to the experimental data,the results show that this method can effectively from the time snapshot of the data found time series trends.