连续属性上的OLAP查询建模方法研究  

Modeling for OLAP Queries on Continuous Dimensions

作　　者：高雅卓[1,2] 倪志伟[1,2] 倪丽萍[1,2] 唐李洋[1,2] 

机构地区：[1]合肥工业大学管理学院智能管理研究所,合肥230009 [2]合肥工业大学过程优化与智能决策教育部重点实验室,合肥230009

出　　处：《情报学报》2011年第4期372-379,共8页Journal of the China Society for Scientific and Technical Information

基　　金：国家863高技术研究发展计划基金项目（2007AA04Z116）; 国家自然科学基金项目（70871033）

摘　　要：本文以支持管理者决策为出发点,为OLAP查询建立统计模型。文章首次将数理统计学中的核密度估计方法及Copula理论相结合引入到OLAP查询建模的研究中,有效地抽取数据立方体的概要知识,在减少数据存储空间的同时,以近似查询的方法实现查询精度与查询时间之间的折衷。该方法的优势在于对连续属性的查询处理,模型的建立使得在连续属性上的查询降低了对物化方体的依赖性,极大地提高了OLAP查询的灵活性。实验分析表明使用该方法可以在保证较高查询精度的条件下大大减少数据立方体的存储空间,加快OLAP查询速度,从而为管理决策提供快速和高效的指导。This paper proposes a statistical model for OLAP query in order to support decision making.Kernel estimation and Copula theory which are conpects in mathematical statistics are first introduced to construct the OLAP query model in this paper.The model can abstract summary knowledge efficiently to answer OLAP queries without accessing the raw data and make a tradeoff between query accuracy and query time while reducing data storage space.OLAP query on continuous dimensions is the prominent advantage of this model,which can reduce the dependence on materilized data cubes while querying on continuous dimensions and increase the flexibility of OLAP query.It＇ s proved in the empirical analysis and experiment result that the space of data cube storage can be greatly reduced and the response rate to OLAP query can be sped up while the higher accuracy is guaranteed so that the model can support the fast and efficient decision making.

关 键 词：OLAP查询 数据立方体 核密度估计 COPULA 分布函数 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]