Hierarchical self-organizing maps for clustering spatiotemporal data

Spatial sciences are confronted with increasing amounts of high-dimensional data. These data commonly exhibit spatial and temporal dimensions. To explore, extract, and generalize inherent patterns in large spatiotemporal data sets, clustering algorithms are indispensable. These clustering algorithms must account for the distinct special properties of space and time to outline meaningful clusters in such data sets. Therefore, this research develops a hierarchical method based on self-organizing maps. The hierarchical architecture permits independent modeling of spatial and temporal dependence. To exemplify the utility of the method, this research uses an artificial data set and a socio-economic data set of the Ostregion, Austria, from the years 1961 to 2001. The results for the artificial data set demonstrate that the proposed method produces meaningful clusters that cannot be achieved when disregarding differences in spatial and temporal dependence. The results for the socio-economic data set show that the proposed method is an effective and powerful tool for analyzing spatiotemporal patterns in a regional context.     

Dasymetric Spatiotemporal Interpolation

This research applies the principles of dasymetric mapping to spatiotemporal interpolation by extending the spatial concepts of zone and area to their temporal analogs of interval and duration, respectively. An example application of dasymetric spatiotemporal interpolation using crime event data is presented. Results indicate that dasymetric spatiotemporal interpolation significantly improves the accuracy of estimates over areal or duration weighting. In addition, even when dasymetric interpolation in either the spatial or temporal dimension is relatively weak, combining dasymetric estimation in both space and time dimensions simultaneously has the potential to amplify the accuracy of the overall dasymetric estimation.     

Housing-market bubble adjustment in coastal communities - A spatial and temporal analysis of housing prices in Midwest Pinellas County,Florida

This study aims to investigate the spatial and temporal dynamics of housing prices associated with the 2007 U.S. housing-market bubble in Midwest Pinellas County, Florida. Two methods were used to examine the spatial and temporal dynamic of price levels: housing characteristic influence estimation and a hedonic modeling approach. Two consistent spatial patterns emerged in the estimated coefficients associated with various housing characteristics, with definitive changes occurring at 600 m and 2200 m from the coast. These changes suggested the existence of three geographical submarkets: a coastal, an intermediate or transitional, and an inland submarket. Three temporal stages were observed from the coefficient trends associated with various housing characteristics: 2002–2005 (bubble-growing), 2006–2008 (bubble-burst), and 2009–2011 (post-bubble). The hedonic models demonstrated a complexity in the determinant of housing prices during the housing peak-bubble period. The inclusion of submarket dummy variables in the hedonic models improved the amount of explained variation. The R² values of the hedonic models from 2002 through 2008, followed by a decrease after the bubble-burst period. The pre-bubble-burst trend in R² suggests that predictable market forces were at work; partly driven by irrational expectations of housing buyers and the perceptions of run-away housing prices during the growth phase of the bubble.     

A method for discovery and analysis of temporal patterns in complex event data

Pattern analysis techniques currently common within geography tend to focus either on characterizing patterns of spatial and/or temporal recurrence of a single event type (e.g., incidence of flu cases) or on comparing sequences of a limited number of event types where relationships between events are already represented in the data (e.g., movement patterns). The availability of large amounts of multivariate spatiotemporal data, however, requires new methods for pattern analysis. Here, we present a technique for finding associations among many different event types where the associations among these varying event types are not explicitly represented in the data or known in advance. This pattern discovery method, known as T-pattern analysis, was first developed within the field of psychology for the purpose of finding patterns in personal interactions. We have adapted and extended the T-pattern method to take the unique characteristics of geographic data into account and implemented it within a geovisualization toolkit for an integrated computational-geovisual environment we call STempo. To demonstrate how T-pattern analysis can be employed in geographic research for discovering patterns in complex spatiotemporal data, we describe a case study featuring events from news reports about Yemen during the Arab Spring of 2011–2012. Using supplementary data from the Global Database of Events, Language, and Tone, we briefly summarize and reference a separate validation study, then evaluate the scalability of the T-pattern approach. We conclude with ideas for further extensions of the T-pattern technique to increase its utility for spatiotemporal analysis.     

Linking Spatial and Temporal Variation at Multiple Scales in a Heterogeneous Landscape*

Anthropogenic, ecological, and land‐surface processes interact in landscapes at multiple spatial and temporal scales to create characteristic patterns. The relationships between temporally and spatially varying processes and patterns are poorly understood because of the lack of spatiotemporal observations of real landscapes over significant stretches of time. We report a new method for observing joint spatiotemporal landscape variation over large areas by analyzing multitemporal Landsat data. We calculate the spatiotemporal variation of the Normalized Difference Vegetation Index (NDVI) in the area covered by one Landsat scene footprint in north central Florida, over spatial windows of 10⁴–10⁸ m² and time steps of two to sixteen years. The correlations, slopes, and intercepts of spatial versus temporal regressions in the real landscape all differ significantly from results obtained using a null model of a randomized landscape. Spatial variances calculated within windows of 10⁵–10⁷ m² had the strongest relationships with temporal variances (regressions with both larger and smaller windows had lower coefficients of determination), and the relationships were stronger with longer time steps. Slopes and y‐intercepts increased with window size and decreased with increased time step. The spatial and temporal scales at which NDVI signals are most strongly related may be the characteristic scales of the processes that most strongly determine landscape patterns. For example, the important time and space windows correspond with areas and timing of fires and tree plantation harvests. Observations of landscape dynamics will be most effective if conducted at the characteristic scales of the processes, and our approach may provide a tool for determining those scales.     

Space–time interaction of residential burglaries in Wuhan,China

Borrowing methods from epidemiology, studies of spatiotemporal regularities of crime have been booming in various industrialized countries. However, few such attempts are empirical studies using crime data in developing countries due to a lack of data availability. Utilizing a recent burglary dataset in Wuhan, the fourth largest city in China, current research applied the sequential kernel density estimation and the space–time K-function methods to analyze the spatiotemporal changes of hotspots of residential burglaries. The results show that, both spatial and spatiotemporal clustering exists. The hotspots were relatively stable over time. The space–time clustering, however, shows significant concentrations both in space and over time. In addition, analytic results show significant effects of distance decay in terms of occurrences of burglary incidents along the spatial and temporal dimensions. Moreover, findings from the research provide critical information on the space–time rhythm of crime, and therefore can be utilized in crime prevention practice. Finally, the implications of the findings and limitations are discussed.     

Leveraging correlation across space and time to interpolate geophysical data via CoKriging

Managing geophysical data generated by emerging spatiotemporal data sources (e.g. geosensor networks) presents a growing challenge to Geographic Information System science. The presence of correlation poses difficulties with respect to traditional spatial data analysis. This paper describes a novel spatiotemporal analytical scheme that allows us to yield a characterization of correlation in geophysical data along the spatial and temporal dimensions. We resort to a multivariate statistical model, namely CoKriging, in order to derive accurate spatiotemporal interpolation models. These predict unknown data by utilizing not only their own geosensor values at the same time, but also information from near past data. We use a window-based computation methodology that leverages the power of temporal correlation in a spatial modeling phase. This is done by also fitting the computed interpolation model to data which may change over time. In an assessment, using various geophysical data sets, we show that the presented algorithm is often able to deal with both spatial and temporal correlations. This helps to gain accuracy during the interpolation phase, compared to spatial and spatiotemporal competitors. Specifically, we evaluate the efficacy of the interpolation phase by using established machine-learning metrics (i.e. root mean squared error, Akaike information criterion and computation time).     

广州地铁三号线对周边住宅价格的时空影响效应

以广州地铁3号线及周边住宅项目为例,综合运用可达性相等理论、比较分析法、hedonic模型和GIS空间分析技术,计算地铁对周边住宅价格的影响范围,实证分析其时空影响效应。空间效应结果表明:①地铁站点离市中心越近,影响范围越小,离城区越远,影响范围越大;②地铁对周边住宅具有明显的增值作用,住宅价格与地铁距离间呈显著的正向关系,距离越远影响效应越小;③分区域来看,地铁3号线对番禺区影响较显著,影响范围内住宅平均增值20.48%,而天河和海珠区影响范围内住宅平均增值8.73%。时间效应方面,地铁规划期对天河区和海珠区的房价影响不明显,对番禺区房价具有明显的正效应;施工期对周边房价的影响为先负向而后变为正向;运营期其正向影响更加明显。     

Geostatistical space–time mapping of house prices using Bayesian maximum entropy

Mapping spatial processes at a small scale is a challenge when observed data are not abundant. The article examines the residential housing market in Fort Worth, Texas, and builds price indices at the inter- and intra-neighborhood levels. To accomplish our objectives, we initially model price variability in the joint space–time continuum. We then use geostatistics to predict and map monthly housing prices across the area of interest over a period of 4 years. For this analysis, we introduce the Bayesian maximum entropy (BME) method into real estate research. We use BME because it rigorously integrates uncertain or secondary soft data, which are needed to build the price indices. The soft data in our analysis are property tax values, which are plentiful, publicly available, and highly correlated with transaction prices. The results demonstrate how the use of the soft data provides the ability to map house prices within a small areal unit such as a subdivision or neighborhood.     

南京城区住宅售租价格时空分异与影响因素

住宅价格空间分异是中国城市地理学和经济地理学近年来关注的热点前沿课题。以南京4560个居住小区为研究总样本,采集2009-2017年间30个季度各小区平均住宅售价和租金,选取6个特征时段和小区分布相对集中的重点研究区,采用克里格插值法分析研究区内住宅售租价格的空间分异与演变特征,发现售价空间分异明显加剧,高值区渐显于河西新城、江心洲和鼓楼名校学区;租金空间则从城市中心向外围递减格局转变为整体更加均衡的新老城区多中心结构。在此基础上,重点围绕住宅“区位”属性,构建售租价格分异影响因素指标体系,通过逐步多元回归分析发现,中心位势变量对售租价格的解释度最高,而配套服务类区位因素对售租价格的解释力在降低。南京城市房价快速增长背景下,常规“区位”因素对房价分异的重要性持续减弱,学区、政策偏向等特殊“区位”因子对房价的决定性作用则逐步突显,而“售租比”全面快速增长则预示着城市房价风险程度的整体提高。     

乡村景观多功能权衡–协同关系时空格局研究——以上海市青浦区为例

景观多功能定量化及其权衡–协同关系研究是景观生态学研究的热点。研究针对景观功能相互表征的科学性和景观多功能间关系定量化问题,以上海市青浦区为例,采用"生产–生活–生态"功能框架构建6种乡村景观功能,利用1980–2018年的基础数据,采用Spearman秩相关系数分析和双变量空间自相关相结合的方法,对该区域184个行政村的乡村景观多功能间权衡–协同关系变化特征进行研究,得到以下结果:(1)都市郊区的农业生产功能不能取代经济发展功能,青浦区经济发展功能已成为主导的功能形态。(2)青浦区乡村景观多功能间权衡–协同关系时空差异显著,表现为:农业生产功能与经济发展功能、经济发展功能与生态调节功能、经济发展功能与环境维衡功能始终呈现为权衡关系,为该区不可回避的矛盾,该结果表明该区经济发展是以牺牲生态环境为代价实现的;经济发展功能与景观美学功能间权衡关系显著,且二者关系格局呈东部权衡、西部协同,权衡区的数量递增;经济发展功能与空间承载功能间关系格局呈现东部权衡和西部协同格局,且协同关系区呈增长趋势;景观美学功能与生态调节功能/环境维衡功能间权衡区集中在青东区域,协同区分布于青西区域,且权衡区数量先增后减,协同区先减后增。(3)时空结合的分析方法能准确全面地反映乡村景观多功能间权衡–协同关系演变特征。     

Graph-assisted landscape monitoring

The structure of computational spatial analysis has mostly built on data lattices inherited from cartography, where visualization of information takes priority over analysis. In these framings, spatial relationships cannot easily be encoded into traditional data lattices. This hinders spatial analysis that emphasizes how interactions among spatial entities reflect mutual inter-relationships. This paper explores how graph theoretic principles can support spatiotemporal analysis by enabling assessment of spatial and temporal relationships in landscape monitoring.     

Immigration,Polarization, or Gentrification? Accounting for Changing House Prices and Dwelling Values in Gateway Cities

Past research has identified immigration, social polarization, and gentrification as factors with significant impacts upon price movements and other housing characteristics in gateway cities. This study attempts to compare the effects of these three factors in Toronto and Vancouver, Canada's primary gateway cities, over the period from 1971 to 1996. The paper describes house price changes from Multiple Listing Service rolls and changes of dwelling values in census tracts, and interprets visual evidence for the effects of the three factors. The observed centralization of price gains is then sharpened in a univariate and multivariate analysis of changes in dwelling values for census tracts in each metropolitan area. While there is consistency in the spatial patterns of changes in housing prices and dwelling values between the two cities, there are differences in the importance of the three processes at different times and places. Moreover, strong effects at the metropolitan scale become much more blurred with spatial disaggregation.     

市域范围下商品住宅价格空间自相关分析——以抚州为例

基于2005～2009年抚州市域范围内普通商品住房价的数据资料,通过计算Moran’s,和LocalMoran＇s,系数值对抚州市域范围内普通住宅房价的时空分布格局进行空间自相关分析。分析结果表明,抚州市域范围下商品住宅房价在整体上未表现出空间自相关性;但在局部范围不同年度间少数城镇存在空间相关性。研究表明,房价空间自相关特性与研究区域的空间范围,经济整体发展水平、交通网络发达程度、区域内购房政策有关。     

基于空间扩展模型和地理加权回归模型的城市住房价格空间分异比较

鉴于已有研究主要集中探讨住房价格的空间依赖性,较少涉及空间异质性对住房特征价格的影响,也很少尝试构建不同计量模型来比较模型间刻画住房价格影响因素空间分异的准确性,以长沙市中心城区为研究区,采用空间扩展模型和地理加权回归模型比较分析城市住房价格影响因素的空间分异,结果表明：① 空间扩展模型和地理加权回归模型都表明,长沙市中心城区的住房属性边际价格随着区位的变化而变化,揭示住房价格影响因素具有显著的空间异质性;小区环境、交通条件、教育配套、生活设施等因素对住房价格的影响强度存在明显的空间分异。② 地理加权回归模型和空间扩展模型都能对传统特征价格模型进行改进,但地理加权回归模型在解释能力和精度方面都超过空间扩展模型;对属性系数估计空间模式的分析,地理加权回归模型形成的结果比采用坐标多义扩展的空间扩展模型更为复杂和直观。     

The Use of Comaps to Explore the Spatial and Temporal Dynamics of Fire Incidents: A Case Study in South Wales,United Kingdom*

This article reports on the results from a spatiotemporal analysis of disaggregate fire incident data. The innovative analysis presented here focuses on the exploration of spatial and temporal patterns for four principal fire incident categories: property, vehicle, secondary fires, and malicious false alarms. This research extends previous work on spatial exploration of spatiotemporal patterns by demonstrating the benefits of comaps and kernel density estimation in examining temporal and spatiotemporal dynamics in calls for services. Results indicate that fire incidents are not static in either time or space and that spatiotemporal variation is related to incident type. The application of these techniques has the potential to inform policy decisions both from a reactive, resource‐allocation perspective and from a more proactive perspective, such as through spatial targeting of preventive measures.     

Exploring spatiotemporal clusters based on extended kernel estimation methods

We examined three different ways to integrate spatial and temporal data in kernel density estimation methods (KDE) to identify space–time clusters of geographic events. Spatial data and time data are typically measured in different units along respective dimensions. Therefore, spatial KDE methods require special extensions when incorporating temporal data to detect spatiotemporal clusters of geographical event. In addition to a real-world data set, we applied the proposed methods to simulated data that were generated through random and normal processes to compare results of different kernel functions. The comparison is based on hit rates and values of a compactness index with considerations of both spatial and temporal attributes of the data. The results show that the spatiotemporal KDE (STKDE) can reach higher hit rates while keeping identified hotspots compact. The implementation of these STKDE methods is tested using the 2012 crime event data in Akron, Ohio, as an example. The results show that STKDE methods reveal new perspectives from the data that go beyond what can be extracted by using the conventional spatial KDE.     

空间分析方法在房地产市场研究中的应用--以北京市为例

基于北京市2003年普通住宅数据,利用空间分析中点模式分析、空间自相关分析和空间插值方法等,对北京市房地产,尤其是普通住宅的空间格局进行分析。研究表明,北京市房地产发展在空间上具有强烈的集聚特点,而房价的空间自相关特性也非常明显。空间分析方法提供了准确认识、评价和综合理解空间位置和空间相互作用的手段,为定量研究空间格局提供了支持。在房地产发展等社会经济现象研究中,空间分析方法强调了“位置”因素的重要性,是刻画房地产空间格局的理想工具。     

基于STARMA模型的城市暴雨积水点积水短时预测

近年来城市暴雨出现突发和多发态势,导致城市内涝灾害频繁发生,威胁着城市居民的生命和财产安全。随着城市降雨积水监测网的建立,获得分钟尺度的降雨和积水时序监测数据成为可能,实现了城市内涝的实时监控。但目前对监测数据的利用仍显不足,缺乏对其深度分析挖掘,造成监测系统“只监不控”的局面。本文基于城市降雨积水监测网的监测数据,根据积水时间相关性、降雨空间相关性以及降雨积水序列相关性,构建降雨积水的时空自相关移动平均模型(STARMA),对城市暴雨积水点积水过程进行短时预测。STARMA模型已被广泛应用于交通预测、环境变量预测以及社会经济领域,特别是在时空过程机理不清楚、多因素时空变量影响的情况下效果较好。本文首次将该模型应用到降水积水过程拟合和积水短时预测上,同时在方法上改进了传统单变量的STARMA模型,建立降雨和积水双变量的STARMA模型模拟降雨积水过程。并以北京市2012年“7.21”事件降雨积水过程为研究对象,以丰北桥、花乡桥、马家楼桥和六里桥4个积水监测点为例,建立降雨积水的STARMA模型,以5 min为步长作积水5、10、15 min三步预测。验证结果表明,该模型在降雨积水过程中拟合效果较好,模型短时预测精度较高。该项研究能够有效地利用监测数据,提高信息预警和应急指挥能力,为市政防汛或交通等部门提供决策支持。     

Fine-grained prediction of urban population using mobile phone location data

Fine-grained prediction of urban population is of great practical significance in many domains that require temporally and spatially detailed population information. However, fine-grained population modeling has been challenging because the urban population is highly dynamic and its mobility pattern is complex in space and time. In this study, we propose a method to predict the population at a large spatiotemporal scale in a city. This method models the temporal dependency of population by estimating the future inflow population with the current inflow pattern and models the spatial correlation of population using an artificial neural network. With a large dataset of mobile phone locations, the model’s prediction error is low and only increases gradually as the temporal prediction granularity increases, and this model is adaptive to sudden changes in population caused by special events.