多尺度空间同位模式挖掘的点过程分解方法

Point process decomposition method for multi-scale spatial co-location pattern mining

空间同位模式挖掘旨在发现多类地理要素或事件频繁互邻近形成的关联规则,对于理解复杂地理现象内在发生机理具有重要价值。针对现有基于关联规则的空间同位模式挖掘方法难以对地理数据特性(如多尺度性)进行有效建模的缺陷,本文提出了一种基于点过程建模与分解的多尺度空间同位模式挖掘方法。首先通过构建一个随机变量将多类要素实例分布数据表达为混合空间点过程,并引入一个非参统计指标对同位模式进行特征尺度判别;基于此,定义一种条件概率密度分布函数,利用点过程分解思想挖掘多尺度空间同位模式及其实例分布。试验分析结果表明本文方法可以准确挖掘空间同位模式在不同尺度的空间分布形态,并且有效降低了人为设定参数的主观性。

Spatial co-location pattern mining aims to discover association rules formed by multiple types of geographic elements or events frequently adjacent to each other,which is the key for understanding the internal occurrence mechanism of complex geographic phenomena.Aiming at the shortcomings of existing spatial colocation pattern mining methods in the effective modeling of geographic data characteristics(such as the multi-scale characteristic),this paper proposes a multi-scale spatial co-location pattern mining method based on point process decomposition.Firstly,the spatial distribution of geographical elements with multiple types is modeled as a mixed spatial point process by constructing a random variable,and a non-parametric statistical index is introduced to discriminate the characteristic scale of the co-location patterns.On this basis,we define a conditional probability density distribution function to mine multi-scale spatial co-location patterns using points process decomposition.The experimental analysis results show that the proposed method can accurately depict the spatial distribution of spatial co-location patterns at different scales,and effectively reduce the subjectivity of artificially setting parameters.