城市功能分区的空间聚类方法研究及其应用——以济南市为例

A Study on Spatial Clustering of Urban Function Partition

作为空间数据挖掘的一种重要手段,空间聚类目前已在许多领域得到了应用,它是城市功能分区中的关键性步骤。根据空间-属性一体化的概念模型,把影响城市功能分区的空间坐标、空间关系和属性特征纳入到统一的空间计算模型,分别运用K-平均算法、神经网络方法,对城市功能分区进行空间聚类计算,充分挖掘空间坐标和空间关系数据中隐含的空间聚集信息。实例分析表明,基于神经网络的空间聚类结果可以为城市功能分区提供准确、可靠的依据。

As an important means of spatial data mining, spatial clustering has been applied in many fields at present. Spatial clustering is the key process of urban function partitioning. Based on the conception model of com- bination of coordinate and attribute, this paper brings spatial coordinate, spatial relationship and attribute features into the unitive model of spatial computation. Urban function is divided adopting means and ANN. This method fully mines the connotative spatial clustering information in spatial attribute data and spatial positions. The experi- ment shows that the unitive spatial clustering method based on ANN can provide a sufficient and reliable basis for urban function partitioning.