顾及缺失值的因果图时空预测网络

时空预测是地理时空大数据挖掘的基础研究命题。目前,多种模型用于预测未知系统的时空状态。然而,存在的大多数预测模型仅在没有缺失数据的时空数据集上进行测试,忽略了缺失值对预测结果的影响。在真实场景中,由于传感器或网络传输故障,数据缺失是一个不容忽视的问题。鉴于此,本文提出了一种顾及缺失值的因果图卷积网络(causal graph convolutional network considering missing values, Causal-GCNM)模型用于时空预测。Causal-GCNM模型可以自动捕捉时空数据中的缺失模式,使得Causal-GCNM模型在不需要借助额外插值算法的前提下,可以直接完成时空预测任务。本文提出的模型在3种真实的时空数据集(交通流数据集、PM_2.5监测数据集及气温监测数据集)得到了验证。试验结果表明,Causal-GCNM模型在4种缺失条件(20%随机缺失、20%块状缺失、40%随机缺失及40%块状缺失)下仍然具有较好的预测性能,并在预测精度和计算效率两类指标上优于10种存在的基线方法。     

网络文本数据时空感知计算概念、方法及应用场景

网络文本数据作为网络传感数据的一种,因其包含大量的时空信息,以及丰富的语义信息,不仅能够用于研究社会问题的时空特征,也能够对社会现象的时空规律提供语义上的可解释性,因此存在一定的应用价值,目前已逐渐得到研究者的关注。本文提出了网络文本数据的一般概念表示,分析了网络文本数据的类型和特征,重点梳理了网络文本数据时空感知计算的计算方法,总结了文本优先+地理背景和地理优先+文本增强两种计算模式,并从静态到动态、场景到区域的视角总结了4个相关应用场景。     

地理空间智能研究进展和面临的若干挑战

随着地理空间科学、人工智能、高性能计算技术的迅速发展,地理空间智能已成为处理和分析地理空间大数据的主要手段,并将在地球科学、空间认知、智慧城市、智慧社会等科学研究、工程建设和社会发展中发挥越来越重要的作用.地理空间智能作为地理空间科学和人工智能深度融合的交叉领域,其发展受到多学科的驱动,目前已在算力增强软硬件研制、系统开发、数据与模型共享、服务与应用方面不断取得进展,显示出巨大的活力和潜能,同时难题和挑战也相生相伴.本文首先阐述地理空间智能的概念演进、若干技术系统构建思路和国内外科学研究现状,然后梳理地理空间智能的典型应用,分析地理空间智能面临的问题和挑战,最后对其重要的发展方向及趋势予以展望.     

基于网络Voronoi图启发式和群智能的最大覆盖空间优化

提出一种基于网络Voronoi面域图的最大覆盖选址模型及相应的粒子群优化方法,并应用于城市响应时间敏感型公共服务设施的空间优化。本文考虑设施功能沿交通网络辐射以及需求非均匀分布情形,对设施在网络连续空间上进行布局优化,选址模型采用网络Voronoi面域图划分布局设施的功能辐射域,以启发空间优化最小化重叠覆盖。模型同时顾及了设施利用率的最大化,规定设施对给定距离以内的需求实行的完全服务覆盖和对给定距离以外的需求实行随距离衰减的部分服务覆盖。本研究提出基于遗传机制和广义Voronoi图改进的粒子群算法以提高其空间优化性能,通过对南京市消防站最大覆盖空间优化实验表明,该研究取得了较为理想的结果,可应用于城市化区域应急设施最大覆盖空间优化。     

基于深度学习的城市PM2.5浓度时空分布预测及不确定性评估

城市PM2.5浓度的时空分布预测旨在基于有限观测样本实现研究区域内PM2.5分布的全范围感知。理想的预测模型需同时保证结果的高精度与高可靠性。然而,现有研究大多以提升精度为唯一目的,忽视了由于数据质量与模型结构的各异所导致预测结果的不确定性,这极大限制了高精度预测结果的可靠性与可用潜力,从而难以有效辅助空气污染治理等实际应用。为此,本文提出一种耦合不确定性评估的PM2.5浓度时空分布预测模型。通过构建以图卷积和循环网络为主的预测模块,实现PM2.5浓度的高精度预测;同时,基于对抗学习策略与变分自编码思想构建不确定性量化模块,同步揭示预测结果的不确定性水平。深圳市实际数据实证表明,本文方法能有效兼顾PM2.5浓度预测结果的精度与可靠性,能为包括监测站点布局选址在内的环境治理工作提供科学决策支持。     

基于互联网的城市时空数据调查技术方法研究

分析互联网时空数据的类型和来源,研究互联网时空数据的获取和挖掘方法。以重庆市空气质量的时空变化研究为例进行案例分析,探讨在时空数据获取和挖掘中的应用。     

基于深度学习的人群活动流量时空预测模型

传统的时空预测方法缺乏对复杂时空非线性关系的描述,且难以顾及空间多尺度特征对于预测结果的影响.针对这一问题,本文提出了一种融合空间多尺度特征的时空网络模型(MST-Net),将流量预测的回归问题转换为具有时空特性的判别模型.首先,通过并联卷积提取空间多尺度特征;然后,通过引入注意力机制的门控循环单元提取时间特征;最后,...     

基于加权图的地籍时空数据描述模型研究

以计算机通用的数据结构——图为基础，研究了基于加权图的时空数据组织策略，为表达时空语义提供了一种形式化的描述方法，并分析了以该模型为中间件实现不同时空模型转换的应用前景。     

基于时空对象的高铁网络建模方法

全空间信息系统将现实世界中的实体抽象为多粒度时空对象,并对其进行表达与分析,较传统地理信息系统来说更符合人类的认知。对高铁网络对象的基本特点及现有交通网络建模方法进行了分析,可知目前现有的建模方法难以支持具有动态性、关联性及多粒度特性的高铁网络对象的表达。依据多粒度时空对象建模思想,提出了一种适合高铁交通数据的全面表达与操作的建模方法;并对高铁网络对象化建模方法的优势及建模过程进行了探讨,设计了概念模型和逻辑模型;通过实例对提出的建模方法进行了验证,并采用可视化表达的方法检验高铁网络时空对象数据的有效性。实验结果表明,基于时空对象的建模方法能够较好描述高铁时空实体多维动态特征,能为后续高铁网络时空查询、分析提供支持。     

人际关系网络对象化时空建模方法研究

以全空间信息系统的多粒度时空对象数据模型为指导思想,进行人际关系网络的对象化时空建模方法研究.首先,分析了人际关系网络的建模内容;然后,设计了面向人际关系网络的多粒度时空对象逻辑数据模型;最后,提出了一种人际关系网络的对象化批量构建方法并进行了实例验证.结果表明,本文所提出的人际关系网络对象化时空建模方法将人际关系网络...     

A graph-based method for Chinese address matching

Address matching is a substantial task in location-based services. Currently, major address matching methods either perform rather badly on unstructured data or fail to extract adequate semantic information of address elements. In this article, we propose a graph-based method that can deal with both sides of the problem. First, we use a pretrained transformer neural network to handle address tokenization. Then we parse address tokens into address elements according to their parts of speech. Then the node2vec and tf-idf technique is used to generate node embeddings for each address element. Finally, an address matching graph convolutional network is applied to do the address matching work. We have carried out a series of experiments on a real-world Chinese address corpus, to further evaluate the impacts of our methods. The experimental results indicate our method achieves higher scores than the state-of-the-art methods.     

Measuring geographic segregation: a graph-based approach

Residential segregation is a multidimensional phenomenon that encompasses several conceptually distinct aspects of geographical separation between populations. While various indices have been developed as a response to different definitions of segregation, the reliance on such single-figure indices could oversimplify the complex, multidimensional phenomena. In this regard, this paper suggests an alternative graph-based approach that provides more detailed information than simple indices: The concentration profile graphically conveys information about how evenly a population group is distributed over the study region, and the spatial proximity profile depicts the degree of clustering across different threshold levels. These graphs can also be summarized into single numbers for comparative purposes, but the interpretation can be more accurate by inspecting the additional information. To demonstrate the use of these methods, the residential patterns of three major ethnic groups in Auckland, namely Māori, Pacific peoples, and Asians, are examined using the 2006 census data.     

一种新的地理本体构建方法分析

对于地理本体的研究是在考虑地理信息自身特殊性的基础上,把本体在信息科学中的含义移植到地理信息科学中,实现对空间信息语义的丰富。目前没有一个标准的地理本体构造方法。本文针对当前地理本体构建方法的不足,提出了一种新的地理本体构建方法。该方法强调需求分析的重要性,采用螺旋式上升的构建过程,构建了海运领域地理本体,实现了海运地理信息的语义共享。     

Cybercity: Conception,technical supports and typical applications

In this paper,the concepts about the CyberCity and the corresponding technique supports and typical applications are introduced.As the pilot projects,the applications of CCGIS software in CyberShenzhen and CyberShanghai are also presented.At last,some of key technologies such as the reconstruction of three dimensional city model,vast amount of data integration and its real time application of 3D animation are discussed.     

Space-Time Intelligence Systems: Technology, applications and methods

Geographic Information System (GIS) software is constrained, to a greater or lesser extent, by a static world view that is not well-suited to the representation of time (Goodchild 2000). Space Time Intelligence System (STIS) software holds the promise of relaxing some of the technological constraints of spatial only GIS, making possible visualization approaches and analysis methods that are appropriate for temporally dynamic geospatial data. This special issue of the Journal of Geographical Systems describes some recent advances in STIS technology and methods, with an emphasis on applications in public health and spatial epidemiology.The STIS expert workshops were funded in part by grants R01CA092669 and R01CA096002 from the National Cancer Institute, and by grants R43-ES010220 and R44-ES010220 from the National Institute of Environmental Health Sciences. Gillian AvRuskin provided cheerful editorial assistance. We thank the participants at the workshops for providing invaluable expertise and critical insights.     

Spatially restricted data distributions on the sphere: the method of orthonormalized functions and applications

 In many geoscientific applications data are irregularly distributed and not globally available, e.g. caps around the poles which are uncovered due to non-polar satellite orbits, or signals being defined solely on bounded regions on the globe. Starting from a sequence of base functions with global support, which in the present case is composed of spherical harmonics being initially non-orthogonal on a bounded subdomain, a set of functions is generated that constitutes an orthonormal basis. Different approaches to realize this transformation are studied and compared with respect to numerical stability and computational effort, and the corresponding effects on the coefficient recovery are investigated. A number of synthetic tests demonstrate the applicability, the benefit, but also the limitations, of this method. Received: 24 March 2000 / Accepted: 9 October 2000     

A graph-based approach for the structural analysis of road and building layouts

A better understanding of the relationship between the structure and functions of urban and suburban spaces is one of the avenues of research still open for geographical information science. The research presented in this paper develops several graph-based metrics whose objective is to characterize some local and global structural properties that reflect the way the overall building layout can be cross-related to the one of the road layout. Such structural properties are modeled as an aggregation of parcels, buildings, and road networks. We introduce several computational measures (Ratio Minimum Distance, Minimum Ratio Minimum Distance, and Metric Compactness) that respectively evaluate the capability for a given road to be connected with the whole road network. These measures reveal emerging sub-network structures and point out differences between less-connective and more-connective parts of the network. Based on these local and global properties derived from the topological and graph-based representation, and on building density metrics, this paper proposes an analysis of road and building layouts at different levels of granularity. The metrics developed are applied to a case study in which the derived properties reveal coherent as well as incoherent neighborhoods that illustrate the potential of the approach and the way buildings and roads can be relatively connected in a given urban environment. Overall, and by integrating the parcels and buildings layouts, this approach complements other previous and related works that mainly retain the configurational structure of the urban network as well as morphological studies whose focus is generally limited to the analysis of the building layout.     

Recent advances in accessibility research: Representation,methodology and applications

 In this article we examine recent advances in accessibility research and their implications for future studies. We base our discussion on three intersecting dimensions that are useful for evaluating the contribution of recent studies: representation, methodology and applications. Various examples are selected to show that research concerned with representation and methodological problem solving is often applied to issues of broad concern in policy and planning. It is, however, not clear that the simultaneous treatment of representation, methodological and application issues has ever been fully worked out. The questions raised in this article may serve as a foundation for addressing issues pertinent to accurate representation, improved model building, and more rigorous applications in accessibility research. Received: 9 December 2002 / Accepted: 10 February 2003     

Crack measurement: Development,testing and applications of an automatic image-based algorithm

The paper presents an Image-based Method for Crack Analysis (IMCA) which is capable of processing a sequence of digital imagery to perform a twofold task: (i) the extraction of crack borders and the evaluation of its width across the longitudinal profile; (ii) the measurement of crack deformations (width, sliding and rotation). Here both problems are solved in 2-D, but an extension to 3-D is also addressed. The equipment needed to apply the method is made up of a digital camera (or a still video-camera in case a high frequency in data acquisition is necessary), an orientation frame which establishes the object reference system, a pair of signalized supports to be placed in a permanent way on both sides of the crack to compute deformations; however, permanent targets are mandatory only for case (ii). The measurement process is carried out in a fully automatic way, a fact also that makes this technique highly operational for unskilled people in engineering surveying or photogrammetry. The accuracy of the proposed method, evaluated in experimental tests adopting different consumer digital cameras, is about ± 5–20 μm, like the accuracy of most deformometers, but with the advantage of automation and of augmented achievable information; moreover, the image sequence can be archived and off-line measurements could be performed at any time.