目前的地理信息系统都使用定量的方法存储和查询空间信息,不能反映人们的定性思维方式,因此它表达地理空间信息的能力受到限制。在关于空间关系的自然语言描述的调查的基础上,本文定义了一组能反映线状物体空间关系的度量指标,结合反映拓扑关系的定量指标,使用决策树的数据挖掘算法,对调查结果进行分析,形式化了描述空间关系的自然语言词汇,建立了模糊的自然语言表达和图形的几何特征之间的联系。自然语言形式化后得到的规则可用于空间关系的自然语言查询系统。

People usually use qualitative terms to express spatial relations,while current geographic information systems(GIS) all use quantitative approaches to store spatial information.The abilities of current GIS to represent and query spatial information about geographic space are limited.If GIS are to be made easier to use,it must be made more intuitive,so that it will help people focus on the work they want to do without thinking about a computer program or application.In order to incorporate the concepts and methods people used to infer information about geographic space into GIS,research on the formal model of common sense geography becomes increasingly important.This paper worked on how to formalize the qualitative natural-language descriptions of spatial relations between linear geographic objects.Based on the result of a human subject test about natural-language description of spatial relations,this paper defines some quantitative topological and metric indices which are related with the natural-language spatial relation terms,and uses these indices to formalize the natural-language representation with a decision tree algorithm,thus bridging the gap between natural-language terms and the computational model of spatial relations.This method provided a way of completely understanding the relationship between the ambiguous natural-language representations and the geometric spatial relations of geographic objects.The rules extracted from the trees can be used in natural-language spatial query systems.