利用城市POI数据提取分层地标

为了获取能够用于智能化路径引导的层次性空间知识,提出了一种依据显著度的差异从城市POI数据中提取出分层地标的方法。首先,通过从公众认知、空间分布和个体特征3个方面分析影响POI显著性的因素,构造了包括公众认知度、城市中心度和特征属性值3个指标向量的POI显著性度量模型;然后,分别讨论了利用问卷调查、多密度空间聚类和数据规格化的方法计算POI对象的各项显著性指标值的过程;最后,选择武汉市武昌地区的POI数据进行显著度计算,从中提取显著度较高的对象构成若干层地标,并以各层地标为种子生成加权的Voronoi图,用来反映各地标的空间影响范围并建立了同层和上下层地标之间蕴含的关系。     

从城市兴趣点中提取多层次地标方法探究

为了获取结构化空间知识,本文在前人研究基础上,提出一种依据显著度差异从城市POI数据中提取多层次地标的方法。分别从知名度、通达度以及个体特征三个方面量化分析了影响POI显著性的因素;然后利用这三个指标构建了POI显著性度量模型,并进行了武汉市武昌区地标提取实验。最后,以各层地标为种子生成Voronoi图,用来反映各层地标的空间影响范围及上下层地标之间蕴含的认知规律。     

微博签到POI数据有效性衡量方法

自发地理信息具有数据量大,信息丰富,成本低廉,现势性强等优势,使得其蕴含巨大的应用价值,但如何有效控制数据质量问题来提升数据的应用价值值得深入研究。本文以微博签到POI数据为研究对象,首先从大众地理空间认知和签到POI数据空间分布两个方面对影响签到POI数据有效性的因素进行分析,然后建立有效性数学衡量的模型提取实验区有效数据。最后通过实验验证了该模型快速、高效提取自发式地理信息方式采集的POI数据可行性。     

基于微博位置签到数据的POI更新方法

POI的现势性对于位置服务至关重要,但传统人工实地调查效率低,现势性无法满足需求。以当前用户参与数众多的微博社交网络为数据平台,提出了一种基于微博位置签到数据的POI更新方法。首先,对微博位置签到数据进行预处理,剔除语义与空间位置不一致的噪声点,在此基础上提出一种基于RANSAC算法的位置签到数据集地理配准方法,实现位置签到数据与已有地理数据库的可靠配准;然后,将位置签到数据集与已有POI数据库进行空间分析与匹配建模,对匹配不成功的位置签到数据进行有效性验证,提取有效新增数据入库用以更新POI;最后,以武汉市的街旁网位置签到数据进行POI更新实验,能够有效地发现新增POI和消失POI,为POI快速高效更新提供了全新的方式。     

基于Python的互联网POI数据爬取

随着互联网的快速发展,兴趣点(POI)数据遍布网络,人们可以找到自己感兴趣的餐馆、电影院、商店等,同时产生了大量利用互联网POI数据进行增值服务开发的业务.然而,面对互联网海量的POI数据,很难通过传统手工下载并整理数据,使得很多针对POI数据开发的业务受阻.鉴于此,以Python为编程语言,以百度地图为平台,构建了互联网POI数据爬取方法框架;并以南昌餐馆POI数据爬取为例进行了实验.结果表明,爬取的POI数据真实有效且效率远高于传统手工调查方法,为POI数据获取提供了良好的方法思路.     

POI与NPP/VIIRS夜光数据空间耦合关系下的城市空间结构分析——以武汉市主城区为例

城市兴趣点（POI）和夜光遥感影像能够直观反映城市社会经济等实体要素的空间分布特征,在城市空间结构研究中发挥着重要作用。本文首先选取长江中游城市群的典型城市代表——武汉市作为研究区,选用研究区2016年POI和NPP/VIIRS夜光遥感数据作为基础研究数据,采用GIS分析工具对POI数据进行了空间核密度分析;然后分别对POI核密度分析结果和NPP/VIIRS夜光遥感数据进行了空间网格化处理;最后采用双因素制图和栅格叠加分析方法对两类数据的空间耦合关系进行了探讨,并在此基础上进一步分析了城市空间结构特征。研究表明,武汉市POI数据和夜光遥感数据的空间耦合性整体较好,空间耦合相一致区域占比为82.15%;但POI数据和夜光遥感数据的空间耦合性在长江沿岸地区也存在部分差异,如硚口区、汉阳区夜光遥感数据和POI数据多以低—中的空间耦合模式为主,而青山区、武昌区和汉口区则多以中—低的空间耦合模式为主。武汉市作为中原城市群的核心城市之一,其城市内部空间结构与长江经济带发展关联密切,通过对POI和夜光遥感数据的空间耦合关系探讨,能够对武汉市空间实体要素的空间结构特征有更加深入的了解。本文结果可为沿江城市内部空间结构的研究提供一种崭新的视角。     

基于多特征相似性的多源POI匹配方法

本文针对多源POI的特征差异性导致同名对象识别难的问题,提出了一种多特征相似性的多源POI匹配方法。兼顾空间与非空间属性,选取名称、位置、地址、分类4个特征进行相似度计算;采用层次分析法对各特征指标进行重要性分析,得到特征权值;根据总相似度对候选匹配对象进行筛选,以确定最终匹配对象。试验结果显示,该方法具有较高的匹配精度,更适用于多源异构POI数据的匹配,可满足多源POI数据的高效匹配需求。     

基于天地图的POI数据采集系统设计与实现

针对传统POI数据采集方式成本高,更新缓慢,属性信息不完整等问题,分析天地图提供的地理信息服务和编程接口,结合网络开发技术提出了基于天地图的POI数据采集思路,开发了原型系统,实现了POI数据的采集与动态更新,对保障POI数据现势性、正确性和完整性具有参考意义。     

多源异构POI融合方法及应用

针对不同来源的POI在数据格式、坐标参考、属性结构等方面存在差异的问题,提出了一种多源异构POI的融合方法。该方法首先将经过实测并质检的公开版地图数据做成一个母库;然后将其他来源的POI数据经过坐标统一后与母库进行空间位置和门址等属性信息的比对,将母库中没有的POI数据融合进去,最终形成内容规整、信息量丰富的POI融合库;最后进行基于空间约束关系的门址挂接,即使每一个POI点与对应的门牌建立关联关系,提升POI库的应用途径。试验结果表明,该方法能够实现多源POI数据的高效融合,与门址进行挂接后,可满足日益增加的LBS和基于地图的"点式"服务（即Location Point）的需求。     

遥感影像与POI数据相结合的城市建成区提取适用性研究——以沈阳市为例

城市建成区的发展状况是地理国情监测的重要内容,本文基于遥感影像数据和POI数据对城市建成区进行提取,针对二者的适用性问题进行了研究。试验以沈阳市为研究区域,在研究区域内选择2016年遥感影像数据和POI数据作为数据源进行对比分析。首先,对遥感影像数据和POI数据进行预处理;其次,通过监督分类的方法对遥感影像进行建成区的提取;然后,采用核密度估计法分析POI数据并提取出建成区;最后,利用叠加分析法对比分析这两种数据的适用性。试验结果表明：使用遥感影像数据作为数据源可以较为全面客观地反映城市建成区的发展现状;利用POI数据提取出的城市建成区具有较强的经济属性,能够很好地反映出城市中的经济活跃区。     

Extracting urban functional regions from points of interest and human activities on location‐based social networks

Data about points of interest (POI) have been widely used in studying urban land use types and for sensing human behavior. However, it is difficult to quantify the correct mix or the spatial relations among different POI types indicative of specific urban functions. In this research, we develop a statistical framework to help discover semantically meaningful topics and functional regions based on the co‐occurrence patterns of POI types. The framework applies the latent Dirichlet allocation (LDA) topic modeling technique and incorporates user check‐in activities on location‐based social networks. Using a large corpus of about 100,000 Foursquare venues and user check‐in behavior in the 10 most populated urban areas of the US, we demonstrate the effectiveness of our proposed methodology by identifying distinctive types of latent topics and, further, by extracting urban functional regions using K‐means clustering and Delaunay triangulation spatial constraints clustering. We show that a region can support multiple functions but with different probabilities, while the same type of functional region can span multiple geographically non‐adjacent locations. Since each region can be modeled as a vector consisting of multinomial topic distributions, similar regions with regard to their thematic topic signatures can be identified. Compared with remote sensing images which mainly uncover the physical landscape of urban environments, our popularity‐based POI topic modeling approach can be seen as a complementary social sensing view on urban space based on human activities.     

城市街区和签到数据结合的个性化城市兴趣区域推荐方法

利用社交媒体用户的历史签到数据分析用户空间活动偏好实现用户兴趣区域推荐,在城市商业规划中起着重要作用,也为了解人们的城市生活和需求提供帮助。已有方法获得的ROI具有模糊性和多样性,无法给ROI赋予准确的地理描述信息,对用户来说可解释性不强。因此,本文提出了一种结合城市街区和签到数据的个性化兴趣区域推荐方法(CBCD),引入城市街区概念解决ROI边界模糊问题。首先,通过城市道路网生成城市街区,并将大规模签到数据映射到城市街区转换为区域签到;然后,基于区域签到对用户空间活动偏好和类别偏好分别进行个性化建模;最后,融合空间和类别活动偏好,向用户推荐其可能感兴趣的区域。在真实的数据集上进行试验,结果表明该方法具有较高的推荐精度,对用户感兴趣城市街区的挖掘和推荐具有一定的价值。     

Exploring and Visualizing Differences in Geographic and Linguistic Web Coverage

This article reports on a study performed to understand the geographic and linguistic coverage of web resources, focusing on the example of tourism‐related themes in Switzerland. Search engine queries of web documents were used to gather counts for phrases in four different languages. The study focused on selected populated places and tourist attractions in Switzerland from three gazetteer datasets: topographic gazetteer data from the Swiss national mapping agency (SwissTopo); POI data from a commercial data provider (Tele Atlas) and user generated geographic content (geonames.org). The web counts illustrate the geographic extent and trends of web coverage of tourism for different languages. Results show that coverage for local languages, i.e. German, French and Italian, is more strongly related to the region of the spoken language. Correlation of the web counts to typical tourism indicators, e.g. population and number of hotel nights rented per year, are also computed and compared.     

大众点评数据下的城市场所范围感知方法

“场所（place）”是具有特定语义和人文体验的地理位置,是虚拟地理环境中地理知识表达的核心要素,能够为深入理解与表达人们的地理空间认知,以及基于虚拟地理环境的分析模拟提供支持,因而对城市内部现有场所的识别与模糊建模是城市空间结构研究的基础之一。众源数据为提取和表达模糊场所提供了新的途径,然而目前研究多针对单个或少数几个场所进行建模。本研究针对城市多场所空间范围的多尺度建模,提出基于自适应核密度估计的模糊集方法,为进一步理解城市场所的模糊认知范围提供了可视化解决方案。并以北京市五环内场所为实例,采用大众点评网中商户自行上传的兴趣点（POI）数据,对其进行提取与表达。通过对比百度地图所展示的场所相应范围,发现该方法在大众点评数据集下的应用能更好地揭示商圈语义下的场所认知。     

Multiple factors influence the consistency of cropland datasets in Africa

Accurate geo-information of cropland is critical for food security strategy development and grain production management, especially in Africa continent where most countries are food-insecure. Over the past decades, a series of African cropland maps have been derived from remotely-sensed data, existing comparison studies have shown that inconsistencies with statistics and discrepancies among these products are considerable. Yet, there is a knowledge gap about the factors that influence their consistency. The aim of this study is thus to estimate the consistency of five widely-used cropland datasets (MODIS Collection 5, GlobCover 2009, GlobeLand30, CCI-LC 2010, and Unified Cropland Layer) in Africa, and to explore the effects of several limiting factors (landscape fragmentation, climate and agricultural management) on spatial consistency. The results show that total cropland area for Africa derived from GlobeLand30 has the best fitness with FAO statistics, followed by MODIS Collection 5. GlobCover 2009, CCI-LC 2010, and Unified Cropland Layer have poor performances as indicated by larger deviations from statistics. In terms of spatial consistency, disagreement is about 37.9 % at continental scale, and the disparate proportion even exceeds 50 % in approximately 1/3 of the countries at national scale. We further found that there is a strong and significant correlation between spatial agreement and cropland fragmentation, suggesting that regions with higher landscape fragmentation generally have larger disparities. It is also noticed that places with better consistency are mainly distributed in regions with favorable natural environments and sufficient agricultural management such as well-developed irrigated technology. Proportions of complete agreement are thus located in favorable climate zones including Hot-summer Mediterranean climate (Csa), Subtropical highland climate (Cwb), and Temperate Mediterranean climate (Csb). The level of complete agreement keeps rising as the proportion of irrigated cropland increases. Spatial agreement among these datasets has the most significant relationship with cropland fragmentation, and a relatively small association with irrigation area, followed by climate conditions. These results can provide some insights into understanding how different factors influence the consistency of cropland datasets, and making an appropriate selection when using these datasets in different regions. We suggest that future cropland mapping activities should put more effort in those regions with significant disagreement in Sub-Saharan Africa.     

一种应用兴趣点数据进行地表覆盖分类的方法

针对传统基于遥感影像的地表覆盖分类方法普遍存在的生产周期长、成本高、自动化程度低等问题,提出了一种完全利用兴趣点（point of interest,POI）数据进行地表覆盖自动化分类的方法。首先应用潜在狄利克雷分布主题计算模型,从POI数据的文本信息中挖掘出与地表覆盖类型相关的主题类型和分布概率;然后基于POI文本的主题分布,运用支持向量机分类算法构建地表覆盖分类模型;最后以遥感影像地表覆盖分类结果为依据,采用随机抽样的方式对所提方法进行验证。结果表明,该方法能够较好地区分人造地表和非人造地表,且整体分类精度超过80%,可作为传统遥感影像分类的辅助手段,满足地表覆盖快速分类的制图需求。     

A weighted multi-attribute method for matching user-generated Points of Interest

To a large degree, the attraction of Big Data lies in the variety of its heterogeneous multi-thematic and multi-dimensional data sources and not merely its volume. To fully exploit this variety, however, requires conflation. This is a two-step process. First, one has to establish identity relations between information entities across different data sources; and second, attribute values have to be merged according to certain procedures that avoid logical contradictions. The first step, also called matching, can be thought of as a weighted combination of common attributes according to some similarity measures. In this work, we propose such a matching based on multiple attributes of Points of Interest (POI) from the Location-based Social Network Foursquare and the local directory service Yelp. While both contain overlapping attributes that can be used for matching, they have specific strengths and weaknesses that make their conflation desirable. For instance, Foursquare offers information about user check-ins to places, while Yelp specializes in user-contributed reviews. We present a weighted multi-attribute matching strategy, evaluate its performance, and discuss application areas that benefit from a successful matching. Finally, we also outline how the established POI matches can be stored as Linked Data on the Semantic Web. Our strategy can automatically match 97% of randomly selected Yelp POI to their corresponding Foursquare entities.     

Extracting central places from the link structure in Wikipedia

Explicit information about places is captured in an increasing number of geospatial datasets. This article presents evidence that relationships between places can also be captured implicitly. It demonstrates that the hierarchy of central places in Germany is reflected in the link structure of the German language edition of Wikipedia. The official upper and middle centers declared, based on German spatial laws, are used as a reference dataset. The characteristics of the link structure around their Wikipedia pages, which link to each other or mention each other, and how often, are used to develop a bottom‐up method for extracting central places from Wikipedia. The method relies solely on the structure and number of links and mentions between the corresponding Wikipedia pages; no spatial information is used in the extraction process. The output of this method shows significant overlap with the official central place structure, especially for the upper centers. The results indicate that real‐world relationships are in fact reflected in the link structure on the web in the case of Wikipedia.