地理计算及其前沿问题  

Geocomputation and Frontier Research

作　　者：王铮[1] 隋文娟[1] 姚梓璇[1] 廖悲雨[1] 吴一平[1] 

机构地区：[1]华东师范大学教育部地理信息重点实验室

出　　处：《地理科学进展》2007年第4期1-10,共10页Progress in Geography

基　　金：中科院知识创新工程重要方向项目(KZCXZ-YW-305-4);国家科技支撑计划(2006BAC07B01)

摘　　要：地理计算是地理信息科学的核心内容之一,主要研究地理信息科学的方法学问题,内容包括建模、算法、计算体系和一般方法学问题。本文介绍了地理计算的五个前沿问题:(1)地学数据挖掘从地理学问题出发,对各种数据作地理学的模型处理和结果计算以发现地理知识;(2)空间运筹在地理学中的应用日益广泛,它的算法更加简单严密、精度也更高;(3)多自主体系统模拟已经成为地理学科学研究中除归纳和演绎之外的第三种重要研究方法;(4)离散空间的定性计算是进行地理空间计算的必要基础;(5)本体论的发展是地理信息科学乃至整个地理学发展的需要。Geocomputation, as one of the cores of geographic information science （GIScience）, focuses on methodological problems, including modeling, algorithms, computational systems, and the general methodology in the field of GIScience. With the rapid development of GIScience in China, five trends of geocomputation are concluded and presented respectively： （1） Geo-data mining refers to the means of discovering geographic knowledge concerning geographical problems in ways of model processing and computation of various kinds of data; （2） Spatial operational research is widely applied to the field of geography, with its algorithms simpler and more precise and its accuracy higher; （3） Muhi-Agent System（MAS） has become the third important method of geography research ranking after induction and deduction; （4） Quantitative computation in discrete space turns into the necessary fundation of computation in geographic space; （5） Geographic ontology is essential to the progress of GIScience as well as the development of geography in general. Some new direction have appeared on geocomputation, they include： 1 computational pattern, as is, Cooperative Computing, Pool Computing; spatial computation complexity; and mappattern recognition.

关 键 词：地理计算 数据挖掘 空间运筹 多自主体系统 离散空间 本体论 

分 类 号：P91[天文地球—自然地理学]