数据挖掘方法在北京市用水规律分析中的应用  

Application of Data Mining Method in analysis of water use principle in Beijing Municipality

作　　者：张小娟[1] 蒋云钟[1] 邢志[1] 甘治国[1] 成建国[2] 

机构地区：[1]中国水利水电科学研究院,北京100044 [2]水利部水文局,北京100053

出　　处：《中国水利》2007年第16期21-23,共3页China Water Resources

基　　金：水利部"948"项目(编号:200514)"SAS数据挖掘套件"。

摘　　要：以北京市为研究区域,利用数据挖掘工具中的偏相关分析法分别得出影响生活、第一产业、第二产业、第三产业用水量的主要因素,通过回归分析方法建立各部门用水量与其主要影响因素之间的函数关系,并针对各用水部门提出了控制用水的措施和建议。通过应用实例表明,相关分析和回归分析作为数据挖掘的最基本方法,可以快速、准确地发现数据之间的相关关系,建立反映客观规律的规则和模式,为用水管理提供决策支持。Taking Beijing Municipality as the point of study, partial correlation method of data mining tool is used for analyzing main factors that affect water uses of domestic, first, second and third industries. Functional relation between water usages of varies users and main influential factor is established by regression method and measure or suggestions on water use limit of water users are proposed. This case demonstrates that correlation analysis and regression analysis are the basic methods of data mining, which can find the relation between data quickly and accurately, give rules and models of reflecting objective law and principle and provide support for decision-making in water use management.

关 键 词：数据挖掘 相关分析 回归分析 用水 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]