【摘要】 空间数据具有海量、非线性、多尺度、高维和模糊性等复杂性特点,空间数据挖掘技术是对空间数据中非显性的知识、空间关系等模式的自动提取。该文从空间数据挖掘的知识类型、方法、体系结构、过程以及与GIS系统集成等方面对其进行综述。重点阐述空间特征及区分规则、空间分类及聚类规则、空间分布及关联规则、空间序列及演化规则等知识类型以及统计分析、机器学习、探索性数据分析、可视化分析等数据挖掘方法。通过对空间数据挖掘理论、应用和系统实现等方面研究方向、存在问题的分析,指出集数据库、知识库、专家系统、决策支持系统、可视化工具、网络等技术于一体的空间数据挖掘系统是其主要发展方向。更多还原

【Abstract】 Spatial data mining discovers and extracts implicit knowledge,spatial relations,spatial rules and other interesting patterns from spatial data which have complexity characters of mass,nonlinearity,multi-scale,high-dimension,fuzzy etc.This paper attempts an annotated review of knowledge type,methods,architecture,process and integration with GIS which are related with the field of spatial data mining,and especially pays attention to knowledge type including spatial characteristic and discriminate rules,spatial classification and clustering rules,spatial distribution and association rules,spatial serial and evolution rules,and at the same time,pays attention to data mining methods including statistical analysis,machine learning,exploratory data analysis,visualization analysis,etc.In addition,the existing problems and research directions of spatial data mining are also analyzed about its theory,application and implement of software system.In conclusion,the main trends of spatial data mining are expected through seamless integration with database,knowledge database,expert system,decision system,visualization,and network.更多还原