技术扩散和资本流动作用下中国区域空间结构演化——基于Agent的模拟

空间结构是区域的基本组成部分,在空间交互过程和区域经济发展方面起着重要的作用。本文的主要研究目标是在不同的道路交通系统作用下,中国区域空间结构的发展演化过程。这对于解决中国当前经济发展中的不平衡问题,探索使区域平衡发展的方法具有重要意义。与传统的研究方法不同,本文从自上而下建模的角度构建了一个基于Agent 的动态区域经济模型。在模型中,中国的每个地级市都被定义为企业Agent,拥有不同技术水平的劳动者Agent 为企业提供劳动力,并在Wilson 空间吸引力模型的作用下在区域之间迁移。技术作用于劳动力之上,随劳动者的迁移而扩散。资本由企业的投资行为而转移。模型中的道路交通线具有不同的权值,这意味着道路交通网是异质的。通过情景模拟,得到以下结论：在分级道路交通网的作用下,传统的中心-腹地扩散模式已不再适用。技术服从枢纽-网络结构从中心城市向外扩散。资本的流动也具有与技术扩散相似的过程和空间结构。交通条件可以在很大程度上影响区域的发展,同时自然条件仍然起着空间锁定的作用。优惠的税率政策可以在一定程度上打破这种空间锁定,有利于吸引劳动者迁移,提高当地的生产技术水平。     

Mapping the popularity of urban restaurants using social media data

Nowadays, geographers show growing interest in providing location-based services for urban residents. It is of great practical significance to screen and recommend the most popular restaurants to consumers, as dining is important to every urban dweller. Consumer review website (CRW) has emerged as an active social media platform in catering industry. This paper demonstrated how to quantify the popularity of urban restaurants (PUR) by using CRW, namely, Dianping.com. An applied popularity index (PI) was developed to quantify PUR, based on the consumer review scores (food, service, and decoration) and physical data (evaluation frequency and restaurant grade) for 8259 restaurants within the Hangzhou city, China. All the information, together with the geographic location data, was harvested from the corresponding Application Programming Interface (API) platform. PUR was then mapped by using a geographic information system (GIS). Results showed that restaurants of high popularity were generally concentrated in old urban districts, whereas those in new urban districts presented low popularity. The kernel density distribution of PI also highlighted the geography that the PI values declined from the central city toward the outskirts. Locational associations between PUR and urban functional units (bank, shopping mall, school, cinema, hotel, bar, and scenic spot) generally presented a similar tendency. Restaurants with high PI values were in high proximity to urban functional units, whereas those with low PI values were located far away from other functional units. It implied that restaurants with high popularity tended to be located in high mixture with urban functional units. Although we used Dianping.com to perform the analysis, the presented methodology can be extended to other types of CRWs. Our study is believed to provide new insights into applied geographic sciences.     

社交媒体数据在自然灾害应急管理中的应用研究综述

社交媒体是灾害应急管理的一个新兴但尚未充分利用的大数据源。然而,社交媒体数据在灾害应急管理中能发挥什么作用,目前国内的研究较少,而国外的研究亦较为分散,相关应用受限。论文对国内外利用社交媒体数据进行灾害管理相关研究进行了综述,归纳出社交媒体数据在灾害应急管理中应用的潜力、优势和问题。结果发现：① 社交媒体数据含有丰富的灾害信息,能够依靠态势感知和信息共享支持灾害预警、实时监测、损失和救助需求评估、协助快速应急响应以及预测可能的灾害风险等灾害管理工作;② 社交媒体数据的优势在于能保证灾害信息的实时性和连续性,进而提供致灾强度图、损失预测和舆情分析等结果服务于灾害应急管理;③ 地理位置信息有限、专业语料库缺乏、信息噪声处理的复杂性等原因使社交媒体数据分析结果精度受限。如何发挥社交媒体数据社会感知的优势,以弥补传统观测调查手段的不足,更加科学应对灾害应急,仍需要在社交媒体数据处理及多源数据融合分析技术方面取得突破。     

Coupling maximum entropy modeling with geotagged social media data to determine the geographic distribution of tourists

Modeling the geographic distribution of tourists at a tourist destination is crucial when it comes to enhancing the destination’s resilience to disasters and crises, as it enables the efficient allocation of limited resources to precise geographic locations. Seldom have existing studies explored the geographic distribution of tourists through understanding the mechanisms behind it. This article proposes to couple maximum entropy modeling with geotagged social media data to determine the geographic distribution of tourists in order to facilitate disaster and crisis management at tourist destinations. As one of the most popular tourist destinations in the United States, San Diego was chosen as the study area to demonstrate the proposed approach. We modeled the tourist geographic distribution in the study area by quantifying the relationship between the distribution and five environmental factors, including land use, land parcel, elevation, distance to the nearest major road and distance to the nearest transit stop. The geographic distribution’s dependency on and sensitivity to the environmental factors were uncovered. The model was subsequently applied to estimate the potential impacts of one simulated tsunami disaster and one simulated traffic breakdown due to crisis events such as a political protest or a fire hazard. As such, the effectiveness of the approach has been demonstrated with specific disaster and crisis scenarios.     

The convergence of GIS and social media: challenges for GIScience

It is hard to believe that 10 years have passed since we wrote our guest editorial for IJGIS (Sui and Goodchild 2001 Sui, D.Z. and Goodchild, M.F. 2001. Are GIS becoming new media?. International Journal of Geographical Information Science, 15(5): 387–390. [Taylor & Francis Online], [Web of Science ®] , [Google Scholar]). Using the nascent evidence that emerged in the late 1990s, we speculated back in 2001 that geographic information systems (GIS) were rapidly becoming part of the mass media. On the basis of the proposition of GIS as media, we were able to link GIScience with theories in media studies such as Marshall McLuhan's law of the media, which considers modern media as modifiable perceptive extensions of human thought (Sui and Goodchild 2003 Sui, D.Z. and Goodchild, M.F. 2003. A tetradic analysis of GIS and society using McLuhan's law of media. Canadian Geographers, 47(1): 5–17. [Crossref], [Web of Science ®] , [Google Scholar]). Remarkable conceptual and technological advances in GIS have been made during the past 10 years. The goal of this review is to provide an update on the ‘GIS as media’ argument we made 10 years ago and to discuss the new challenges for GIScience posed by the growing convergence of GIS and social media.     

Geosocial gauge: a system prototype for knowledge discovery from social media

The remarkable success of online social media sites marks a shift in the way people connect and share information. Much of this information now contains some form of geographical content because of the proliferation of location-aware devices, thus fostering the emergence of geosocial media – a new type of user-generated geospatial information. Through geosocial media we are able, for the first time, to observe human activities in scales and resolutions that were so far unavailable. Furthermore, the wide spectrum of social media data and service types provides a multitude of perspectives on real-world activities and happenings, thus opening new frontiers in geosocial knowledge discovery. However, gleaning knowledge from geosocial media is a challenging task, as they tend to be unstructured and thematically diverse. To address these challenges, this article presents a system prototype for harvesting, processing, modeling, and integrating heterogeneous social media feeds towards the generation of geosocial knowledge. Our article addresses primarily two key components of this system prototype: a novel data model for heterogeneous social media feeds and a corresponding general system architecture. We present these key components and demonstrate their implementation in our system prototype, GeoSocial Gauge.     

Interaction between scale and scheduling choices in simulations of spatial agents

Spatial simulations are a valuable tool in understanding dynamic spatial processes. In developing these simulations, it is often required to make decisions about how to represent features in the environment and how events unfold in time. These spatial and temporal choices have been shown to significantly alter model outcomes, yet their interaction is less well understood. In this paper, we make use of a simple group foraging model and systematically vary how features are represented (cell size of the landscape) as well as how events unfold in time (order in which foragers take action) to better understand their interaction. Our results show similar nonlinear responses to changes in spatial representation found in the literature, and an effect of the order in which agents were processed. There was also a clear interaction between how features are represented and how events unfold in time, where, under certain environmental representations results were found to be more sensitive to the order in which individuals were processed. Furthermore, the effects of feature representation, scheduling of agents, and their interaction were all found to be influenced by the heterogeneity of the spatial surface (food), suggesting that the statistical properties of the underlying spatial variable will additionally play a role. We suggest that navigating these interactions can be facilitated through a better understanding of how these choices affect the decision landscape(s) on which agents operate. Specifically, how changes to representation affect aggregation and resolution of the decision surface, and thereby the degree to which agents interact directly or indirectly. We suggest that the challenges of dealing with spatial representation, scheduling, and their interaction, while building models could also present an opportunity. As explicitly including alternate representations and scheduling choices during model selection can aid in identifying optimal agent–environment representations. Potentially leading to improved insights into the relationships between spatial processes and the environments in which they occur.     

Classifying urban land use by integrating remote sensing and social media data

Urban land use information plays an important role in urban management, government policy-making, and population activity monitoring. However, the accurate classification of urban functional zones is challenging due to the complexity of urban systems. Many studies have focused on urban land use classification by considering features that are extracted from either high spatial resolution (HSR) remote sensing images or social media data, but few studies consider both features due to the lack of available models. In our study, we propose a novel scene classification framework to identify dominant urban land use type at the level of traffic analysis zone by integrating probabilistic topic models and support vector machine. A land use word dictionary inside the framework was built by fusing natural–physical features from HSR images and socioeconomic semantic features from multisource social media data. In addition to comparing with manual interpretation data, we designed several experiments to test the land use classification accuracy of our proposed model with different combinations of previously acquired semantic features. The classification results (overall accuracy = 0.865, Kappa = 0.828) demonstrate the effectiveness of our strategy that blends features extracted from multisource geospatial data as semantic features to train the classification model. This method can be applied to help urban planners analyze fine urban structures and monitor urban land use changes, and additional data from multiple sources will be blended into this proposed framework in the future.     

中国大都市多中心空间演化过程的非均衡动态模拟

构建了一个基于垄断竞争、规模经济、空间成本、消费者多样化偏好以及商品和服务差异化的大都市多中心空间结构模型,并整合基于agent的建模,借助非均衡(out-of-equilibrium)动态模拟,探讨了人口规模和消费需求、区域间和区域内商业产品及服务的替代弹性、固定成本投入、区位和交通条件等因素的"持续"变化对的大都市空间结构演化的影响过程。模拟表明:以商业为代表的经济活动往往集聚在具有区位优势、规模经济和消费能力的大都市传统中心,而交通条件的持续改善、新兴郊区次中心与传统城市中心之间以及城区内部商业产品和服务的不断差异化对大都市的多中心空间结构有重要影响,但影响的程度不同。更重要的是,大都市的人口规模和消费能力对新兴郊区次中心的形成作用显著。     

Airline networks and the international diffusion of severe acute respiratory syndrome (SARS)

In fewer than four months in 2003, severe acute respiratory syndrome (SARS) spread from China to 25 countries and Taiwan, becoming the first new, easily transmissible infectious disease of the twenty-first century. The role of air transport in the diffusion of the disease became obvious early in the crisis; to assess that role more carefully, this study relates the spatial-temporal pattern of the SARS outbreak to a measure of airline network accessibility. Specifically, the accessibility from those countries that were infected by SARS, beginning with China, to other countries was measured using airline schedules. The country-pair accessibility measure, along with other country-level factors relevant to the disease, were tested as determinants of the speed with which SARS arrived in infected countries as well as its failure to arrive in most countries. The analyses indicate that airline network accessibility was an especially influential variable but also that the importance of this variable diminished in the latter weeks of the outbreak. The latter finding is partly attributable to public health measures, particularly health screening in airports. The timing and geography of those measures are reviewed using data from media reports and interim World Health Organization (WHO) documents during the outbreak. The uneven effort to curtail the international diffusion of SARS suggests further planning is needed to develop a concerted response to contain future epidemics.     

Location‐based social media behavior and perception: Views of university students

Location‐based social media (LBSM), a specific type of volunteered geographic information (VGI), is increasingly being used as a spatial data source for researchers in geography and related disciplines. Many questions, though, have been raised about VGI data in terms of its quality and its contributors. While a number of studies have explored users’ demographics and motivations for contribution to explicitly geographic forms of VGI, such as OpenStreetMap and Wikimapia, few have focused on these aspects with implicitly geographic forms of VGI, such as LBSM (for example, Twitter and Instagram). This study, through use of an online survey, specifically assesses the LBSM behavior and perceptions of 253 university students, noting differences found in gender, race, and academic standing. We find that the greatest differences are those between males and females, rather than through race or academic standing, and LBSM appears less biased than other forms of VGI.     

ASSURE: a model for the simulation of urban expansion and intra-urban social segregation

Numerous cities in developing regions worldwide are expanding at a tremendous rate. This requires adequate strategies to address the needs of these growing cities with diverse populations. Nonetheless, the development of urban policies is often hampered by the lack of reliable data or insight in the socio-spatial dynamics of this urban expansion. This paper therefore presents ASSURE, a spatially and temporally explicit model that can simulate urban growth and intra-urban social segregation, taking into account alternative policy strategies and expected social dynamics. The model has a flexible structure that allows incorporating specific city conditions that influence residential decision-making and adapting the simulation to the data available. This, in combination with the transparent model structure, makes ASSURE a potentially valuable decision support tool for urban planning. The potential is demonstrated with an example where the urban growth of and social segregation in Kampala (Uganda) is simulated based on (semi-)quantitative and qualitative data for ca. 800 households collected through interviews. The results of the simulations show that depending on the scenario, the spatial segregation and accessibility problems will evolve highly differently.     

Mapping spatiotemporal patterns of events using social media: a case study of influenza trends

Tracking spatial and temporal trends of events (e.g. disease outbreaks and natural disasters) is important for situation awareness and timely response. Social media, with increasing popularity, provide an effective way to collect event-related data from massive populations and thus a significant opportunity to dynamically monitor events as they emerge and evolve. While existing research has demonstrated the value of social media as sensors in event detection, estimating potential time spans and influenced areas of an event from social media remains challenging. Challenges include the unstable volumes of available data, the spatial heterogeneity of event activities and social media data, and the data sparsity. This paper describes a systematic approach to detecting potential spatiotemporal patterns of events by resolving these challenges through several interrelated strategies: using kernel density estimation for smoothed social media intensity surfaces; utilizing event-unrelated social media posts to help map relative event prevalence; and normalizing event indicators based on historical fluctuation. This approach generates event indicator maps and significance maps explaining spatiotemporal variations of event prevalence to identify space-time regions with potentially abnormal event activities. The approach has been applied to detect influenza activity patterns in the conterminous US using Twitter data. A set of experiments demonstrated that our approach produces high-resolution influenza activity maps that could be explained by available ground truth data.     

环境友好型农业技术扩散的时空演化与影响因素——基于社会网络视角

面对严峻的农业面源污染问题,环境友好型农业技术的推广和落地对于中国实现乡村生态振兴和农业农村现代化具有重要意义。但现实中,该类型技术在乡村地区一直面临推广难,采纳度低的问题。以往众多研究证明,社会网络是技术扩散的有效途径和支撑。基于此,本文从社会网络的视角构建了农业技术创新扩散的理论框架,通过2016—2020年广东省农业面源污染治理的实践调研及10个县市3015份问卷数据的定量分析,总结出环境友好型农业技术扩散的四个阶段：① 初始阶段：农户间简单离散的技术互动。② 单核阶段：精英农户的点式嵌入与拟亲缘网络扩散。③ 多核阶段：多个精英农户形成与业缘关系网络扩散。④ 高水平互动阶段：农业经济组织的形成与功能性网络扩散。个体资源禀赋、技术有效性及感知度、技术传播方式和外部环境作为技术扩散的四个主要影响因素,在不同的阶段发挥着不同的作用。本研究成果有助于理解中国农业技术扩散的实践逻辑和底层机制,对同类技术推广有重要的政策意义。     

Integrating social network data into GISystems

Today, online social media outlets provide new and plentiful sources of data on social networks (SNs) and location-based social networks (LBSNs), i.e., geolocated evidence of connections between individuals. While SNs have been used to show how the magnitude of social connectivity decreases with distance, there are few examples of how to include SNs as layers in a GISystem. If SNs, and thus, interpersonal relationships, could be analyzed in a geographic information system (GIS) setting, we could better model how humans socialize, share information, and form social groups within the complex geographic landscape.

Our goal is to facilitate a guide for analyzing SNs (as derived from online social media, telecommunications, surveys, etc.) within geographic space by combining the mature fields of social network analysis (SNA) and GISystems. First, we describe why modeling socialization in geographic space is essential for understanding human behavior. We then outline best practices and techniques for embedding SN nodes and edges in GISystems by introducing terms like ‘social flow’ and ‘anthrospace’, and categorizations for data and spatial aggregation types. Finally, we explore case study vignettes of SNA within GISystems from diverse regions located in Bolivia, China, Côte d’Ivoire, Singapore, the United Kingdom, and the United States, using concepts such as geolocated dyads, ego–alter relationships, node feature roles, modularity, and network transitivity.     

A hybrid ensemble learning method for tourist route recommendations based on geo-tagged social networks

Geo-tagged travel photos on social networks often contain location data such as points of interest (POIs), and also users’ travel preferences. In this paper, we propose a hybrid ensemble learning method, BAyes-Knn, that predicts personalized tourist routes for travelers by mining their geographical preferences from these location-tagged data. Our method trains two types of base classifiers to jointly predict the next travel destination: (1) The K-nearest neighbor (KNN) classifier quantifies users’ location history, weather condition, temperature and seasonality and uses a feature-weighted distance model to predict a user’s personalized interests in an unvisited location. (2) A Bayes classifier introduces a smooth kernel function to estimate a-priori probabilities of features and then combines these probabilities to predict a user’s latent interests in a location. All the outcomes from these subclassifiers are merged into one final prediction result by using the Borda count voting method. We evaluated our method on geo-tagged Flickr photos and Beijing weather data collected from 1 January 2005 to 1 July 2016. The results demonstrated that our ensemble approach outperformed 12 other baseline models. In addition, the results showed that our framework has better prediction accuracy than do context-aware significant travel-sequence-patterns recommendations and frequent travel-sequence patterns.     

RecNet: a deep neural network for personalized POI recommendation in location-based social networks

How to exploit various features of users and points of interest (POIs) for accurate POI recommendation is important in location-based social networks (LBSNs). In this paper, a novel POI recommendation framework, named RecNet, is proposed, which is developed based on a deep neural network (DNN) to incorporate various features in LBSNs and learn their joint influence on user behavior. More specifically, co-visiting, geographical and categorical influences in LBSNs are exploited to alleviate the data sparsity issue in POI recommendation and are converted to feature vector representations of POIs and users via feature embedding. Moreover, the embedded POIs and users are fed into a DNN pairwise to adaptively learn high-order interactions between features. Our method is evaluated on two publicly available LBSNs datasets and experimental results show that RecNet outperforms state-of-the-art algorithms for POI recommendation.     

国内外社会空间分异测度研究综述

社会空间分异作为一个具有社会性及空间性的概念,其测度研究直接关系到隔离等级划分的合理性和有效性。本文根据不同历史时期空间分异测度研究所关注的视角,将社会空间分异测度研究分为早期萌芽阶段、多维度群体评估阶段、空间转向阶段、多模型多视角阶段等4个阶段,介绍了社会空间分异测度指标在各阶段的演变特点和主要测度指标,并对其中26个主要指数进行比较和分类,阐述各指数的适用范围和优缺点。分析国内社会空间分异相关研究现状,指出其存在多维度指标使用较少且多为实证性研究,空间分异测算方法缺少理论基础的不足。移动互联网和社交网络空间快速发展,对社会空间分异的测算方法提出新的要求;基于地理位置信息的个人行为空间交互数据和基于虚拟网络的赛博空间交互数据,为社会空间分异研究提供了新的研究视角。