空间层次聚类显著性判别的重排检验方法

A Permutation Test for Identifying Significant Clusters in Spatial Dataset

同时顾及空间邻近与专题属性相似的空间层次聚类是挖掘空间分布模式的一种有效手段。空间层次聚类方法虽然可以获得多层次的聚集结构,但聚类结果显著性的统计判别依然是一个尚未解决的难题。为此,本文提出了一种空间层次聚类结果显著性的统计判别方法,用于确定空间层次聚类的停止准则,减少聚类过程对参数设置的依赖。通过试验分析与比较发现,该方法能够有效判别空间层次聚类结果的显著性和确定层次聚类合并过程的停止条件,同时具有很好的抗噪性,避免随机结构的干扰。

Spatial hierarchical clustering methods considering both spatial proximity and attribute similarity play an important role in exploratory spatial data analysis.Although existing methods are able to detect multi-scale homogeneous spatial contiguous clusters,the significance of these clusters cannot be evaluated in an objective way.In this study,apermutation test was developed to determine the significance of clusters discovered by spatial hierarchical clustering methods.Experiments on both simulated and meteorological datasets show that the proposed permutation test is effective for determining significant clustering structures from spatial datasets.