出租车轨迹数据的南京人群出行模式挖掘

针对现有出租车轨迹数据挖掘中时间序列邻近度量方法存在的问题,提出一种基于DBSCAN算法和改进的DTW距离的时间序列聚类算法提取具有相似性出行特征的时空模式,进而研究城市人群出行行为的时空差异。以南京市为例,结合电子地图对出行模式的空间分布特征进行分析,证明了本文所提出的方法的有效性。实验结果表明:在空间分布上,工作日出租车出行模式按照平均出行频次由高到低排序,从城市中心向四周扩散,呈中心环状分布,出行模式区域界限较为明显,同类出行模式分布区域对应相似的功能。提出了一种基于DBSCAN算法和改进的DTW距离的时间序列聚类算法提取具有相似性出行特征的时空模式,有效地分析城市人群出行行为的时空差异。     

Hydrogeochemical zonation for groundwater management in the area with diversified geological and land-use setup

Despite its limited aerial extent, the National Capital Territory (NCT) Delhi, India, has diversified geological and topographical setup. A geochemical assessment of prevailing conditions of aquifer underlying the NCT was attempted and further classified into different hydrogeochemical zones on the basis of statistical and analyses and its correlation with land use, geological and climatic setting. Mineral phase study and isotopic analyses were used for the verification of performed clustering. Saturation indices (SI) calculated using the geochemical modelling code PHREEQC were used to distinguish the characteristics of four zones, as saturation states of the water does not change abruptly. Four different hydrogeochemical zones were statistically identified in the area: (1) intermediate (land-use-change-impacted) recharge zone, (2) discharge (agriculture-impacted) zone, (3) recharge (ridge) zone, and (4) recharge floodplain (untreated-discharge-impacted) zone. The distinctiveness of hydro-geochemical zones was further verified using stable isotopic (²H and ¹⁸O) signature of these waters. GIS-based flow regime in association with long-term geochemical evidences implied that these zones are being affected by different problems; thus, it necessitates separate environmental measures for their management and conservation. The study suggested that in a diversified urban setup where the complex interactions between anthropogenic activities and normal geochemical processes are functioning, hydrogeochmical zoning based on the integration of various techniques could be the first step towards sketching out the groundwater management plan.     

基于RBF神经网络的城市需水量预测

介绍了径向基函数神经网络的原理、训练算法,建立的径向基函数神经网络城市需水量预测模型具有较强的非线性处理能力和逼近能力,运算速度快、性能稳定,克服了BP神经网络学习过程的收敛过分依赖于初值和可能出现局部收敛的缺陷,预测精度较高,泛化能力强。     

一种利用像素邻域信息的模糊聚类图像分割算法

本文提出一种结合空间信息的模糊C均值聚类图像分割算法。该方法是利用每个像素的邻域像素的隶属度来修正FCM算法的隶属度函数,从而引入图像的空间信息,对隶属函数做了改进;依据平方误差和最小准则,从而确定模糊分类矩阵及聚类中心;并依据最大隶属度原则,划分图像像素的类别归属。实验结果表明,该方法能快速有效地分割图像,并且具有较强的抗噪能力。     

DCAD: a dual clustering algorithm for distributed spatial databases

Spatial objects have two types of attributes: geometrical attributes and non-geometrical attributes, which belong to two different attribute domains (geometrical and non-geometrical domains). Although geometrically scattered in a geometrical domain, spatial objects may be similar to each other in a non-geometrical domain. Most existing clustering algorithms group spatial datasets into different compact regions in a geometrical domain without considering the aspect of a non-geometrical domain. However, many application scenarios require clustering results in which a cluster has not only high proximity in a geometrical domain, but also high similarity in a non-geometrical domain. This means constraints are imposed on the clustering goal from both geometrical and non-geometrical domains simultaneously. Such a clustering problem is called dual clustering. As distributed clustering applications become more and more popular, it is necessary to tackle the dual clustering problem in distributed databases. The DCAD algorithm is proposed to solve this problem. DCAD consists of two levels of clustering: local clustering and global clustering. First, clustering is conducted at each local site with a local clustering algorithm, and the features of local clusters are extracted. Second, local features from each site are sent to a central site where global clustering is obtained based on those features. Experiments on both artificial and real spatial datasets show that DCAD is effective and efficient.     

Characterizing and understanding the variability of streamflow drought indicators within the USA

ABSTRACT

Low streamflow conditions can have adverse consequences for society and river ecology. The variability and drivers of streamflow drought indicators within the USA were investigated using observed streamflow records from 603 gauges across the USA. The analysis was based on two main approaches: (i) low-flow magnitude indicators, and (ii) streamflow deficit indicators. First, we examined how streamflow drought indicators vary spatially across the USA. Second, we used a data-driven clustering method to identify spatial clusters for each indicator. Finally, we assessed the association with regional climate drivers. The results show that the spatial variability of low-flow magnitude indicators is significantly different from the deficit indicators. Further, our clustering approach identifies regions of spatial homogeneity, which can be linked to the extreme regional climate drivers and land–atmosphere interactions. The influence of regional climate on streamflow drought indicators varies more between clusters than between indicators.     

Revealing hidden persistence in maximum rainfall records

ABSTRACT

Clustering of extremes is critical for hydrological design and risk management and challenges the popular assumption of independence of extremes. We investigate the links between clustering of extremes and long-term persistence, else Hurst-Kolmogorov (HK) dynamics, in the parent process exploring the possibility of inferring the latter from the former. We find that (a) identifiability of persistence from maxima depends foremost on the choice of the threshold for extremes, the skewness and kurtosis of the parent process, and less on sample size; and (b) existing indices for inferring dependence from series of extremes are biased downward when applied to non-Gaussian processes. We devise a probabilistic index based on the probability of occurrence of peak-over-threshold events across multiple scales, which can reveal clustering, linking it to the persistence of the parent process. Its application shows that rainfall extremes may exhibit noteworthy departures from independence and consistency with an HK model.     

Scale-specific automated line simplification by vertex clustering on a hexagonal tessellation

A new method of cartographic line simplification is presented. Regular hexagonal tessellations are used to sample lines for simplification, where hexagon width, reflecting sampling fidelity, is varied in proportion to target scale and drawing resolution. Tesserae constitute loci at which new sets of vertices are defined by vertex clustering quantization, and these vertices are used to compose simplified lines retaining only visually resolvable detail at target scale. Hexagon scaling is informed by the Nyquist–Shannon sampling theorem. The hexagonal quantization algorithm is also compared to an implementation of the Li–Openshaw raster-vector algorithm, which undertakes a similar process using square raster cells. Lines produced by either algorithm using like tessera widths are compared for fidelity to the original line in two ways: Hausdorff distances to the original lines are statistically analyzed, and simplified lines are presented against input lines for visual inspection. Results show that hexagonal quantization offers advantages over square tessellations for vertex clustering line simplification in that simplified lines are significantly less displaced from input lines. Visual inspection suggests lines produced by hexagonal quantization retain informative geographical shapes for greater differences in scale than do those produced by quantization in square cells. This study yields a scale-specific cartographic line simplification algorithm, following Li and Openshaw's natural principle, which is readily applicable to cartographic linework. Open-source Java code implementing the hexagonal quantization algorithm is available online.     

Cities On the Plains: The Evolution of Urban Kansas

The process of congressional redistricting, delineating boundaries for districts in which voters elect members to the U.S. House of Representatives, has always been an expensive and controversial process. Congressional districts (CDs) are redrawn due to changes in population reflected by the decennial census to ensure equal representation. Laws and regulations literature identifies eight criteria that may be considered when determining the boundaries of CDs and this article focuses on one of those criteria, maintaining communities of interest (COIs). This criterion requires states to preserve these boundaries when delineating CDs but fails to define a COI. This research proposes and evaluates two approaches to define a COI and examines the extent to which this criterion has been adhered to. One definition uses Thiessen polygons and census designated places to delineate COIs based on known cultural places, whereas the other definition uses cluster analysis to group together people with similar sociodemographic characteristics. The results show that the two definitions are feasible for defining a COI. Furthermore, the states largely maintain the COI boundaries within their CDs by only splitting, at most, 17.1 percent of the COIs defined. Existing literature shows that maintaining COIs within CDs leads to higher voter participation and engagement, as well as better representation. The results show that if either definition was adopted, states could comply with this criterion with relative ease. Furthermore, a standard definition could help reduce the cost and controversy surrounding the redistricting process.     

Regionalization with dynamically constrained agglomerative clustering and partitioning (REDCAP)

Regionalization is to divide a large set of spatial objects into a number of spatially contiguous regions while optimizing an objective function, which is normally a homogeneity (or heterogeneity) measure of the derived regions. This research proposes and evaluates a family of six hierarchical regionalization methods. The six methods are based on three agglomerative clustering approaches, including the single linkage, average linkage (ALK), and the complete linkage (CLK), each of which is constrained with spatial contiguity in two different ways (i.e. the first‐order constraining and the full‐order constraining). It is discovered that both the Full‐Order‐CLK and the Full‐Order‐ALK methods significantly outperform existing methods across four quality evaluations: the total heterogeneity, region size balance, internal variation, and the preservation of data distribution. Moreover, the proposed algorithms are efficient and can find the solution in O(n ²log n) time. With such data scalability, for the first time it is possible to effectively regionalize large data sets that have 10 000 or more spatial objects. A detailed comparison and evaluation of the six methods are carried out with the 2004 US presidential election data.