空间数据立方体的物化视图选择方法研究  

The Research on Materialization View Selection Method of Spatial Data Cube

作　　者：樊博[1] 李一军[1] 

机构地区：[1]哈尔滨工业大学管理学院,哈尔滨150001

出　　处：《高技术通讯》2003年第9期16-20,共5页Chinese High Technology Letters

基　　金：863计划(2001AA136010);国家自然科学基金(70171013)资助项目。

摘　　要：针对决策支持系统(DSS)中集成空间分析能力的应用趋势,研究基于空间数据仓库的一种决策分析工具——空间在线分析处理(Spatial OLAP),拟解决影响空间OLAP在线响应的瓶颈难点——空间数据立方体的物化问题。首先系统地提出空间OLAP的模型,然后对现有空间度量物化视图选择方法进行改进,提出了双向空间Greedy算法。实验证明,该算法在降低选择时间和求解质量两方面具有更好的表现。The integration of spatial data and decision support system(DSS) is urgently needed to provides spatial analytical function and visualization to user. This article investigates a decision analytical tool based on spatial data warehouse--on-line analytical processing (OLAP) and mainly solves the serious problem of slow response time in spatial data cube operation by improving materialization view selection method on spatial measures. The model of spatial OLAP is put forward systematically. Then a new method named double spatial greedy is poposed, which has the advantages of deducing view selection time as well as improving the quality of view selection result.

关 键 词：数据立方体 物化视图选择 决策支持系统 空间OLAP 模型 Greedy算法 格 数据仓库 

分 类 号：P208[天文地球—地图制图学与地理信息工程]