中文摘要：

在数字城市的基础上，研究具有空间分析能力的客户关系管理系统，提出基于空间数据分类挖掘的客户细分方法．从数据模型、数据过滤、分类挖掘算法这三个阶段入手，设计了全新的CRM知识发现体系．首先提出了客户分类的数据模型，设计全新的空间谓词的过滤方法；然后，本文针对空间信息系统所特有的不确定性，设计基于不完备信息系统的粗糙集算法，以实现合理的空间数据分类；最后，提出了客户空间数据分类的软，硬分类规则，更好地指导CRM策略的制定．

英文摘要：

This paper mainly aims to integrate spatially analytical function into CRM. customer segmentation method is presented based on spatial data classification algorithm, and a new knowledge discovery system is designed which includes customers＇data modeling, spatial data filtering and spatial classification algorithm . Firstly, a series of data preparation methods of spatial data classification including the spatial data cube modeling and the filtering method of spatial predicates are explored. Secondly, the rough set theory is used to design spatial data classification algorithm to meet the need of incomplete information in the system. Thirdly, hard rules and soft rules of spatial data classification to are proposed better guide the CRM strategy.