基于GIS与SDM技术的可视化空间数据分类研究

提出将GIS与可视化空间数据挖掘技术之集成的基本框架。在此基础上,基于VisualC++6.0和Ma-pObject2.0组件技术设计和开发了一个可视化交互空间数据挖掘分类系统,系统采用决策树方法和贝叶斯网络作为数据挖掘方法的基本算法,采用训练与学习相结合实现空间数据的分类。文中用实例数据对系统性能、算法和规则有效性进行了验证。结果表明,该系统是一个适用的、可扩展的可视化交互空间数据挖掘工具,系统能够实现数据挖掘实时动态的交互控制,实现了数据挖掘过程的可视化、挖掘模型的可视化和结果的可视化显示、可视化思考、可视化分析与评价。

Study on Spatial Data Classification Based on Integration Between GIS and Spatial Data Mining

A brief introduction of GIS and spatial data mining as well as their integration are proposed.Then a spatial data mining prototype system used to classify spatial data based on Visual C 6.0 and MapObject2.0 are designed and developed,which use Decision Tree and Bayesian Networks as its basic algorithm to realize spatial data mining,and integrate training and learning to classify spatial data.Some test has been done to verify the capability of the system and the validity of these algorithms and rules.The result indicates that this prototype system is a practical and extensible Visual Interactive Spatial Data Mining tool.It can realize the interactive control of data mining dynamically and real-timely.At the same time,the process,models and result of data mining are displayed,thought and analyzed as well as evaluated visually.