【作者】 潘玲；

【导师】 杨春金；

【作者基本信息】 武汉理工大学 ， 信号与信息处理， 2007， 硕士

【摘要】 地理信息系统和数据挖掘是当今信息技术中的两颗璀璨明珠。前者侧重于信息的管理，后者侧重于信息的处理与分析，有着紧密地联系性和互补性。GIS数据库中含有丰富的数据和信息，其中隐含着许多有价值的知识，而目前的GIS系统主要局限于实现数据的录入、查询、统计等功能，无法有效地发现数据中存在的关系和规则，而数据挖掘技术可以对GIS数据进行更高层次的分析，发现其中隐含的知识。因而从GIS的空间数据库中进行知识发现即空间数据挖掘，已成为数据挖掘领域中一个重要的研究方向。近年来两种技术的发展，使数据挖掘和GIS结合以挖掘GIS所管理的海量空间数据背后的知识规律成为可能。如何发现在大型空间数据库中所隐藏的、预先未知的信息以辅助相应的应用，这就是目前空间数据挖掘的任务。同时空间关联规则是空间数据挖掘结果的一种最主要的知识规则，它侧重于确定数据中不同领域之间的联系，找出满足给定支持度和可信度阈值的多个域之间的依赖关系。本文主要通过对知识发现和数据挖掘的研究入手，并结合GIS空间数据库，对空间数据挖掘与GIS的集成模式进行深入研究，并建立相应知识模型，提出了基于MapX的空间数据挖掘体系结构，进行模型数据预处理，实现了关联规则挖掘算法Apriori，由挖掘出的知识规则可以体现出空间数据挖掘与GIS集成后得广泛应用前景。本论文主要工作如下：(1)对知识发现和数据挖掘进行了系统深入的研究，将空间数据挖掘与GIS空间数据库进行联合，对两者的集成模式进行深入研究，并建立GIS数据库知识发现模型；(2)对空间数据的处理和分析过程进行了研究，提出了几种空间数据预处理的方法及GIS中的关联规则挖掘，分析了关联规则算法，并进行实现；(3)结合GIS组件MapX，提出了基于MapX的空间数据挖掘体系结构，实现了关联规则挖掘算法Apriori，建立挖掘算法Apriori试验分析平台，通过分析试验体现出空间数据挖掘与GIS集成后得广泛应用前景。更多还原

【Abstract】 Nowadays, Geographic Information System and Data Mining and Knowledge Discovery technology are two bright pearls of information technology .The former emphasizes particularly on the management of information and the latter emphasizes particularly on the processing and analysis of information.There are abundant data and information in the database of Geographic Information System. These data and information include much implicit and valuable knowledge. Presently GIS is limited to the collection, query and statistic of data. It can not discover relations and rules in data while data mining techniques can analyze GIS data at a higher level which can discover the implicit knowledge. Discovering knowledge from spatial database of GIS with data mining techniques, namely spatial data mining, has become an important research aspect in data mining fields.The development of the two techniques makes it possible to mine the knowledge and rules that behind the massive spatial data which is stored in GIS by integrating the two techniques.The main task of spatial data mining is discovering the implicit, unknown information to assist relevant application. At the same time, spatial association rule is one of the most primary knowledge rules in the result of spatial data mining. It emphasizes particularly on confirming the relationship of data in different fields, and finding out the dependence of data in multi-fields.This paper mainly researches on spatial data mining and GIS Integration mode with GIS spatial database through studing knowledge discovery and data mining.In the paper a association knowledge model and a spatial data mining system architecture based on the MapX are presented. Moreover, model data are pretreated and association rules mining algorithm Apriori is implemented. The knowledge rules by data mining reflect a broad application prospects by integrating spatial data mining and GIS. The main thesis is as follows:(1) Through the research of the knowledge discovery and data mining, combined the spatial data mining and GIS spatial database, and built the knowledge discovery model through in-depth analysis of the spatial data mining and GIS spatial database’s integration modes.(2) Studying the process of data processing and analysing advanced several spatial data reprocessing methods and spatial association rules mining, analyzed and realized the algorithm of spatial association rules.(3) Presenting a spatial data mining architecture based on GIS module MapX , implementing association rules mining algorithm Apriori and establishing mining algorithm Apriori analysis platfoim.it reflects a broad application prospects by integrating the spatial data mining and GIS .更多还原