Spatial analysis using clustering methods: Evaluating central point and median approaches

Spatial analysis is an important area of research which continues to make major contributions to the exploratory capabilities of geographical information systems. The use and application of classic clustering methods is being pursued as an exploratory approach for the analysis of spatially referenced data. Numerous potential clustering approaches exist, so research assessing the relative differences of these approaches is important. This paper evaluates the median and central points optimization based clustering approaches for use in the context of exploratory spatial data analysis. Functional and visual comparisons using three spatial applications across a range of cluster values are carried out. The empirical results suggest that these two clustering approaches identify very similar groupings. The significance of this finding is that the development of clustering tools for exploratory analysis may be limited to the median based approach given relative computational and solvability considerations. Received: 28 September 1998/Accepted: 9 August 1999     

Non-stationary spatial covariance structure estimation in oversampled domains by cluster differences scaling with spatial constraints

In the analysis of spatiotemporal processes underlying environmental studies, the estimation of the non-stationary spatial covariance structure is a well known issue in which multidimensional scaling (MDS) provides an important methodological approach (Sampson and Guttorp in J Am Stat Assoc 87:108–119, 1992). It is also well known that approximating dispersion by a non-metric MDS procedure offers, in general, low precision when accurate differences in spatial dispersion are needed for interpolation purposes, specially if a low dimensional configuration is employed besides a high number of stations in oversampled domains. This paper presents a modification, consisting of including geographical spatial constraints, of Heiser and Groenen’s (Psychometrika 62:63–83, 1997) cluster differences scaling algorithm by which not the original stations but the cluster centres can be represented, while the stations and clusters retain their spatial relationships. A decomposition of the sum of squared dissimilarities into contributions from several sources of variation can be employed for an exploratory diagnosis of the model. Real data are analyzed and differences between several cluster-MDS strategies are discussed.     

德兴铜矿大坞河流域土壤Cu元素形态的聚类分析

在研究重金属元素对德兴铜矿地区环境污染情况的过程中,采用了连续提取法对低品位矿石堆浸场等地的土壤样品进行测量,并以Cu元素存在形态作为指标,综合德兴铜矿地区的特点对Cu元素形态进行了深入的聚类分析研究,得出了Cu在当地土壤中主要以有机态和硅酸盐态存在,但是碳酸盐态等其他四种形态所含比例较大,并且在酸性或氧化环境下具有一定的迁移转化能力,对环境具有潜在危害的结论,为进一步研究其他重金属的迁移转化提供了参考,并为矿区的环境治理和复垦提供了有价值的科学依据.     

多时间尺度密度聚类算法的案事件分析应用

时空聚类是数据挖掘研究的主要内容之一,在环境保护、疾病预防与控制、犯罪预防与打击等领域具有重要的应用价值。已有的时空聚类方法中,时间“距离”都认为是真实的间隔,而对于具有社会属性的案事件而言,其在不同时间尺度下具有明显的周期性特征,忽略这些特征将很难反映出案事件真实的时空规律。本文综合考虑多时间尺度下的时间属性,构建等效时空邻近域,并借鉴经典的密度聚类算法,提出了多时间尺度等效时空邻近域密度聚类算法（MTS-ESTN DBSCAN）。通过对福州市区2013年案事件数据的聚类分析表明,该方法在案事件时空聚类方面具有可行性,对于进一步深入研究城市犯罪地理具有一定的理论意义和实际价值。     

基于局部聚类的网络Voronoi图生成方法研究

提出一种将网络约束下的Voronoi和空间聚类相结合的方法,通过构造局部的聚类分析方法对网络边进行加权,根据实际的点过程性质可以把权重定义为加权或者乘权,进行标准化后与道路段本身长度融合进行计算,依此生成网络Voronoi图,以期理解城市街道的空间特性。以武汉市江汉区为例,对城市网格管理系统产生的城市事件进行算法验证,结果表明,该方法提供了一种灵活的网络约束下的服务区域划分工具,可用于基于网络空间点过程影响下的服务区划分,也可用于系统性地定量刻画城市管理的动态特性。     

基于激光测量和FKM聚类算法的隧洞岩体结构面的模糊群聚分析

岩体隧洞结构面的数量、产状和发育情况是评价隧洞透水性和稳定性的重要指标。采用三维激光测量技术可以快速获取隧洞周围岩体激光点云数据和结构面的几何信息。提出了基于柱面投影和Delaunay生长算法的激光点云的三角网重构技术,利用构建的三角网平面方程的法向向量计算出单个三角网平面的产状,并应用FKM聚类算法进行隧洞岩体结构面产状的模糊群聚分类。该方法可以快速得到不同组的优势产状,且分类结果可用不同颜色进行空间显示。最后运用自主研发的FSS软件对某隧洞的岩体结构面优势产状进行了研究,研究结果表明：1）基于柱面投影和Delaunay生长法的点云构网方法特别适合于隧洞工程,构网速度快、效率高,重构得到的三角网表面与被采样的岩体表面拓扑差别最小;2）FKM模糊群聚方法对隧洞岩体结构面的群聚分类结果可靠、合理,具有较强的通用性和推广性。     

阳澄湖入湖河道分类、污染特征分析及治理策略

为进一步揭示阳澄湖入湖河道的污染物来源,提出相应的治理对策,以2017—2021年阳澄湖入湖河道水质监测数据为基础进行分析讨论。依据入湖水量选取10条主要入湖河道进行分析,其中位于阳澄湖东岸的白曲港在七浦塘拓浚工程建成之前主导流向为出湖,工程建成后,通过Pearson相关分析证实了其流向与七浦塘引水时的水力关系,因此白曲港被选为主要入湖河道。采用距平系数法、系统聚类法和物元分析法将阳澄湖主要入湖河道分为3个类别:第1类别包括白荡、蠡塘河、北河泾、永昌泾4条河道,第2类别包括渭泾塘、界泾和施家斗港3条河道,第3类别包括南消泾、七浦塘和白曲港3条河道。使用因子分析法进行因子分析,第1类别河道的因子为高锰酸盐指数(CODMn)、氨氮(NH3-N)、溶解氧(DO)和总磷(TP),第2类别河道的污染因子为NH3-N、总氮(TN)、pH、TP和DO,第3类别河道的污染因子为pH、TP、TN和DO。通过对上游河道水质情况分析、文件研究以及实地调查等方式得出第1类别河道区域的污染源主要为工业污染源和生活污染源,第2类别河道区域污染源主要为工业污染源与种植业污染源,第3类别河道污染源主要为陆地水产养殖污...     

一种改进的多光谱影像模糊加权聚类方法

多光谱影像具有丰富的地物信息,利用多光谱影像进行聚类分析可以充分利用遥感影像为在大范围地区进行变化检测和土地利用分析作出应有的贡献。而模糊聚类更接近实际情况,能提高影像分类精度,缩短分类时间,提高系统运行效率。本文采用对多光谱影像先进行K-L变换去除了大量的冗余数据后,再进行改进的加权模糊聚类处理方法,取得了较理想的分类效果。     

AN IMPROVED ALGORITHM FOR SUPERVISED FUZZY C-MEANS CLUSTERING OF REMOTELY SENSED DATA

l IntroductionClassification pIays an imPOrtant role for rernotelysensed data tO be intngrated into gapraphical infOr-mation systems(GISs), and is increasingly comPut-eriZed with soPhisticated hardware and software(Cambell l987; Lillesand and Kiefer l994). Pnd-ucts Of classification are usua[ly represented in formof contiguous patches of pixels,with each being la-belled as belonging to a discrete and dominantclass. Such tyPe of classification is termed as crispor discrete. The accuracie…     

A test for the coincident economic and spatial clustering of business enterprises

This paper uses input-output data combined with point process modeling techniques to test whether enterprises linked within nominal buyer-supplier chains have a greater propensity to cluster in space than manufacturing enterprises in general. The methodology controls for the general tendency of firms to seek locations in concentrated agglomerations and isolates the influence of firm interdependence on spatial clustering. Our findings suggest that there is indeed an association between economic linkages and geographic clustering in our study area, but only for some types of economic clusters, mainly those that are comprised mainly of more knowledge-based or technology-intensive sectors. In general, we endeavor to show that spatial analytical methods hold considerable promise for conducting rigorous tests of industrial location questions. Received: 9 September 1998 / Accepted: 12 December 1999