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.     

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.     

A geographic approach for combining social media and authoritative data towards identifying useful information for disaster management

In recent years, social media emerged as a potential resource to improve the management of crisis situations such as disasters triggered by natural hazards. Although there is a growing research body concerned with the analysis of the usage of social media during disasters, most previous work has concentrated on using social media as a stand-alone information source, whereas its combination with other information sources holds a still underexplored potential. This article presents an approach to enhance the identification of relevant messages from social media that relies upon the relations between georeferenced social media messages as Volunteered Geographic Information and geographic features of flood phenomena as derived from authoritative data (sensor data, hydrological data and digital elevation models). We apply this approach to examine the micro-blogging text messages of the Twitter platform (tweets) produced during the River Elbe Flood of June 2013 in Germany. This is performed by means of a statistical analysis aimed at identifying general spatial patterns in the occurrence of flood-related tweets that may be associated with proximity to and severity of flood events. The results show that messages near (up to 10 km) to severely flooded areas have a much higher probability of being related to floods. In this manner, we conclude that the geographic approach proposed here provides a reliable quantitative indicator of the usefulness of messages from social media by leveraging the existing knowledge about natural hazards such as floods, thus being valuable for disaster management in both crisis response and preventive monitoring.     

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.     

STLP-GSM: a method to predict future locations of individuals based on geotagged social media data

ABSTRACT

An increasing number of social media users are becoming used to disseminate activities through geotagged posts. The massive available geotagged posts enable collections of users’ footprints over time and offer effective opportunities for mobility prediction. Using geotagged posts for spatio-temporal prediction of future location, however, is challenging. Previous studies either focus on next-place prediction or rely on dense data sources such as GPS data. Introduced in this article is a novel method for future location prediction of individuals based on geotagged social media data. This method employs the hierarchical density-based clustering algorithm with adaptive parameter selection to identify the regions frequently visited by a social media user. A multi-feature weighted Bayesian model is then developed to forecast users’ spatio-temporal locations by combining multiple factors affecting human mobility patterns. Further, an updating strategy is designed to efficiently adjust, over time, the proposed model to the dynamics in users’ mobility patterns. Based on two real-life datasets, the proposed approach outperforms a state-of-the-art method in prediction accuracy by up to 5.34% and 3.30%. Tests show prediction reliability is high with quality predictions, but low in the identification of erroneous locations.     

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.     

A Local Example of Land‐Use Change: Southern Illinois—1807, 1938, and 1993*

This article provides an analysis of a wetland site in southern Illinois from presettlement to the present. The study area is part of the Cache River‐Cypress Creek Wetland, which has international importance, as recognized by the Ramsar Convention on Wetlands. Land‐cover data for 1807, 1938, and 1993 were created and analyzed with a geographic information system (GIS). Land‐use change by topographic setting (uplands, transitional, and bottomlands) and soil productivity was quantified and studied. Interviews with local experts informed this analysis. Results illustrate the complexity of environmental change and its driving forces. First, notable forest and swamp acreage was converted to cropland between 1807 and 1938 and, to a lesser degree, from 1938 to 1993. Second, there were land‐use variations by topographic region. Between 1807 and 1938, the largest transformation occurred in the uplands, with substantial acreage converted from forest to cropland. Between 1938 and 1993, however, agriculture decreased in the upland areas as hilly areas reverted to forest cover. At the same time, agriculture expanded in the bottomlands as this land was drained for farming. Third, there are interesting patterns within these categories of land‐use change, as soil productivity is an indicator of what lands were taken out of cropland and converted back to grassland and forest.     

Learning about Feminism in Digital Spaces: online methodologies and participatory mapping

Collaborative online research offers opportunities and constraints for geographers. This article critically appraises a collaborative research process that we used to illuminate spatial and political dynamics of feminism contained within the online group ‘Destroy the Joint’ (DTJ). A mostly Australian initiative of over 74 000 Facebook members, DTJ aims to end sexism and misogyny in multiple ways. It operates as a meeting place, discussion forum, and umbrella organisation for numerous micro-campaigns that change in response to broader social, cultural and political contexts, and occur in online and offline spaces. We formed a collaborative research agreement with the moderators of DTJ to reflect on its work, asking participants to put themselves on a map and complete a survey. Participatory GIS and survey-based research operated in the real and the more-than-real spaces that contain paradoxical possibilities. We use the term more-than-real to highlight the excesses of digital spaces: the affect that social media generates, and is generated by, characterises the more-than-real, where extremes in productive and corrosive relations can permeate. Survey results showed diverse appreciations of DTJ's multiple activist tactics (across seemingly ‘superficial’ and ‘meaningful’ interventions), and creative ideas for future campaigns. Mapping revealed both dispersed networks and urban activist concentrations. Nevertheless, after an initial peak of interest and enthusiasm for reflecting on DTJ, the reporting back of research findings to DTJ followers did not garner significant interest. This article tracks that collaborative research trajectory. Intentions to build a hybrid research collective were not realised because of how the more-than-real affords possibilities and limitations. The politics of ‘giving-back’ within the more-than-real are critically engaged with in this appraisal of an experimental online collaboration.     

Coupling mobile phone and social media data: a new approach to understanding urban functions and diurnal patterns

Understanding urban functions and their relationships with human activities has great implications for smart and sustainable urban development. In this study, we present a novel approach to uncovering urban functions by aggregating human activities inferred from mobile phone positioning and social media data. First, the homes and workplaces (of travelers) are estimated from mobile phone positioning data to annotate the activities conducted at these locations. The remaining activities (such as shopping, schooling, transportation, recreation and entertainment) are labeled using a hidden Markov model with social knowledge learned from social media check-in data over a lengthy period. By aggregating identified human activities, hourly urban functions are inferred, and the diurnal dynamics of those functions are revealed. An empirical analysis was conducted for the case of Shenzhen, China. The results indicate that the proposed approach can capture citywide dynamics of both human activities and urban functions. It also suggests that although many urban areas have been officially labeled with a single land-use type, they may provide different functions over time depending on the types and range of human activities. The study demonstrates that combining different data on human activities could yield an improved understanding of urban functions, which would benefit short-term urban decision-making and long-term urban policy making.     

Accessibility to food retailers and socio‐economic deprivation in urban New Zealand

The spatial locations of food retailers are considered to be an influential aspect of population consumption patterns. Such contextual relationships are often related to socio‐economic deprivation, with disparities in accessibility having important implications. This study used Geographic Information Systems and an Enhanced Two‐Step Floating Catchment Area model of spatial accessibility to further understand such relationships within urban areas of New Zealand. Findings, while mixed, indicate that there is generally increased accessibility to all food retailers in highly deprived areas. Understanding these socio–spatial relationships in local environments has important implications for policy initiatives, health outcomes and sustainable development.     

Getting Away with the Stolen Vehicle: An Investigation of Journey‐after‐Crime*

Analyses of criminals' travel patterns can provide significant suggestions to improve crime management. This study extends the investigation of criminals' travel behavior from journey‐to‐crime to journey‐after‐crime. Moreover, new methods are developed to examine the spatial patterns of location pairs when restricted by the underlying geographical process. The methods are employed to investigate criminals' journey‐after‐auto‐theft in the city of Buffalo, New York. The analyses reveal that auto thieves' trips from vehicle‐theft locations to the corresponding vehicle‐recovery locations are local in nature. The travel distances are significantly shorter than the randomly simulated trips; the travel directions are biased from the random directions as well.     

Using open‐ended geographic information system assessments to allow students to construct relevant geographies given the internationalisation of tertiary education in New Zealand

The geospatial skills shortage in New Zealand requires the development of more undergraduate geographic information system (GIS) courses. However, the internationalisation of New Zealand's tertiary education system has resulted in an increasingly diverse tertiary student population, which makes it challenging to teach GIS in a way that maximises relevance to all tertiary students. One approach to this challenge is to make use of the recent proliferation of openly available GIS data, and to internationalise GIS curricula by using open‐ended assessments that provide students with the opportunity to learn GIS by constructing their own geographies of relevance.     

Evaluating alluvial stratigraphic response to cyclic and non‐cyclic upstream forcing through process‐based alluvial architecture modelling

Formation of alluvial stratigraphy is controlled by autogenic processes that mix their imprints with allogenic forcing. In some alluvial successions, sedimentary cycles have been linked to astronomically‐driven, cyclic climate changes. However, it remains challenging to define how such cyclic allogenic forcing leads to sedimentary cycles when it continuously occurs in concert with autogenic forcing. Accordingly, we evaluate the impact of cyclic and non‐cyclic upstream forcing on alluvial stratigraphy through a process‐based alluvial architecture model, the Karssenberg and Bridge (2008) model (KB08). The KB08 model depicts diffusion‐based sediment transport, erosion and deposition within a network of channel belts and associated floodplains, with river avulsion dependent on lateral floodplain gradient, flood magnitude and frequency, and stochastic components. We find cyclic alluvial stratigraphic patterns to occur when there is cyclicity in the ratio of sediment supply over water discharge (Q_s/Q_w ratio), in the precondition that the allogenic forcing has sufficiently large amplitudes and long, but not very long, wavelengths, depending on inherent properties of the modelled basin (e.g. basin subsidence, size, and slope). Each alluvial stratigraphic cycle consists of two phases: an aggradation phase characterized by rapid sedimentation due to frequent channel shifting and a non‐deposition phase characterized by channel belt stability and, depending on Q_s/Q_w amplitudes, incision. Larger Q_s/Q_w ratio amplitudes contribute to weaker downstream signal shredding by stochastic components in the model. Floodplain topographic differences are found to be compensated by autogenic dynamics at certain compensational timescales in fully autogenic runs, while the presence of allogenic forcing clearly impacts the compensational stacking patterns.     

Small‐Scale Enterprise Livelihoods and Social Capital in Eastern Indonesia: Ethnic Embeddedness and Exclusion*

This article explores the importance of specific forms of social capital for small‐scale enterprises while highlighting that such analyses must incorporate local sociocultural complexities. In Eastern Indonesia, small‐scale entrepreneurs in Makassar city rely heavily on informal networks, linkages, and trust relationships for their livelihoods. This dependence reflects different social capital forms, embedded in local ethnic and social relations that are inclusionary for some, yet exclusionary for others. Findings show that although bonding social capital is prevalent, albeit with diverse implications, bridging social capital is less so, and linking social capital is virtually absent. A lack of the latter, combined with widespread corruption in the city, hinders livelihood progress for many local entrepreneurs.     

Health Seeking Behavior: Q‐Structures of Rural and Urban Women in India with Sexually Transmitted Diseases and Reproductive Tract Infections

I use a gender framework to examine why women need to utilize social networks to discover and resolve problems related to reproductive tract infections (RTI) and sexually transmitted diseases (STDs). In India, women's health is linked with social status and perception of their health needs. This paper provides insight on how social networks are gender‐, location‐, and context‐specific and how they are linked to STD issues. This paper examines why social networks are integral to women. The health‐seeking behavior of rural and urban casual sex workers is examined via Q‐analysis, a language of mapping relationships. Results from Q‐analysis reveal women's use of four systems—kinship, belief, traditional medical, and western medical services. Finally, I conclude with important implications for research on gender relations and social networks.     

基于SEM的重大事件国内游客感知价值及行为意向关系研究——2010上海世博会为例

基于上海世博会国内游客调研数据分析,构建世界级博览会游客感知价值维度与行为意向、重游意向的结构方程模型,并进行关系验证.研究表明:(1)上海世博会国内游客感知价值包括服务价值、美感价值、效用价值、愉悦价值、感知价格、便利价值六个测量维度.(2)效用价值、服务价值、愉悦价值对游客行为意向有显著的正向影响,而美感价值、感知...     

新冠肺炎疫情前后游客出行行为的时空特征研究——以中国香港为例

新型冠状病毒肺炎疫情对香港旅游业产生了巨大的冲击,掌握疫情期间游客出行的时空变化特征,对后疫情时代旅游业复苏和发展至关重要。本文基于2019年和2020年社交媒体的旅游评论数据,采用时序统计、复杂网络分析等方法,对比分析疫情前后香港旅游网络的演化特征,并结合旅游目的地语义刻画游客出行规律。结果表明：(1)游客出行规律呈现出一定程度变化,与疫情前相比,游客在一次出游行程中所访问的旅游目的地数量大幅下降;(2)热门目的地仍然在香港旅游网络中承担着连接其他目的地的重要作用,而围绕热门目的地形成的旅游网络社区由疫情前的空间显著集聚转为相对稀疏分布,但网络社区内部目的地的空间分布较疫情前变得更加集聚;(3)在游客选择出行目的地方面,热门目的地访问量急剧减少,而以往游客较少且地理位置较为偏僻的目的地,在游客出行目的地选择时变得更被关注,但产业目的地始终是香港旅游中最热门的旅游目的地类型。     

Sediment dispersal and redistributive processes in axial and transverse deep‐time source‐to‐sink systems of marine rift basins: Dampier Sub‐basin,Northwest Shelf,Australia

Morphological scaling relationships between source‐to‐sink segments have been widely explored in modern settings, however, deep‐time systems remain difficult to assess due to limited preservation of drainage basins and difficulty in quantifying complex processes that impact sediment dispersals. Integration of core, well‐logs and 3‐D seismic data across the Dampier Sub‐basin, Northwest Shelf of Australia, enables a complete deep‐time source‐to‐sink study from the footwall (Rankin Platform) catchment to the hanging wall (Kendrew Trough) depositional systems in a Jurassic late syn‐rift succession. Hydrological analysis identifies 24 drainage basins on the J50.0 (Tithonian) erosional surface, which are delimited into six drainage domains confined by NNE‐SSW trending grabens and their horsts, with drainage domain areas ranging between 29 and 156 km². Drainage outlets of these drainage domains are well preserved along the Rankin Fault System scarp, with cross‐sectional areas ranging from 0.08 to 0.31 km². Corresponding to the six drainage domains, sedimentological and geomorphological analysis identifies six transverse submarine fan complexes developing in the Kendrew Trough, ranging in areas from 43 to 193 km². Seismic geomorphological analysis reveals over 90‐km‐long, slightly sinuous axial turbidity channels, developing in the lower topography of the Kendrew Trough which erodes toe parts of transverse submarine fan complexes. Positive scaling relationships exist between drainage outlet spacing and drainage basin length, and drainage outlet cross‐sectional area and drainage basin area, which indicates the geometry of drainage outlets can provide important constraints on source area dimensions in deep‐time source‐to‐sink studies. The broadly negative bias of fan area to drainage basin area ratios indicates net sediment losses in submarine fan complexes caused by axial turbidity current erosion. Source‐to‐sink sediment balance studies must be done with full evaluating of adjacent source‐to‐sink systems to delineate fans and their associated up‐dip drainages, to achieve an accurate tectonic and sedimentologic picture of deep‐time basins.     

Post‐orogenic thermal history and exhumation of the northern Appalachian Basin: Low‐temperature thermochronologic constraints

Apatite fission‐track (AFT) thermochronology and (U‐Th)/He (AHe) dating, combined with paleothermometers and independent geologic constraints, are used to model the thermal history of Devonian Catskill delta wedge strata. The timing and rates of cooling determines the likely post‐orogenic exhumation history of the northern Appalachian Foreland Basin (NAB) in New York and Pennsylvania. AFT ages generally young from west to east, decreasing from ~185 to 120 Ma. AHe single‐grain ages range from ~188 to 116 Ma. Models show that this part of the Appalachian foreland basin experienced a non‐uniform, multi‐stage cooling history. Cooling rates vary over time, ~1–2 °C/Myr in the Early Jurassic to Early Cretaceous, ~0.15–0.25 °C/Myr from the Early Cretaceous to Late Cenozoic, and ~1–2 °C/Myr beginning in the Miocene. Our results from the Mesozoic are broadly consistent with earlier studies, but with the integration of multiple thermochronometers and multi‐kinetic annealing algorithms in newer inverse thermal modeling programs, we constrain a Late Cenozoic increase in cooling which had been previously enigmatic in eastern U.S. low‐temperature thermochronology datasets. Multi‐stage cooling and exhumation of the NAB is driven by post‐orogenic basin inversion and catchment drainage reorganization, in response to changes in base level due to rifting, plus isostatic and dynamic topographic processes modified by flexure over the long (~200 Myr) post‐orogenic period. This study compliments other regional exhumation data‐sets, while constraining the timing of post‐orogenic cooling and exhumation in the NAB and contributing important insights on the post‐orogenic development and inversion of foreland basins along passive margins.     

基于社会交换理论的社区旅游支持度模型及应用——以黄山风景区门户社区为例

门户社区毗邻风景区且处于其出入口位置,在旅游发展中面临特殊的机遇和难题.依据社会交换理论,选择“社区参与度”、“居住区条件”、“对旅游机构信任度”、“旅游利益感知”、“旅游成本感知”、“社区满意度”和“旅游支持度”7 个结构变量构建社区旅游支持度测度模型,以黄山风景区的4 个门户社区为研究案例,对该模型进行检验,通过结构方程模型分析和因子分析,探讨模型内各变量对社区旅游支持度发挥作用的路径及造成同一风景区不同门户社区支持度差异的影响因素.研究结果表明:① 总体而言,旅游利益感知和社区满意度对旅游支持度有正向影响;居住区条件、对旅游机构信任度和社区满意度对旅游利益感知有正向影响;社区参与度、居住区条件和对旅游机构信任度对社区满意度有正向影响.社区参与度对旅游利益感知正向影响不显著,旅游成本感知这一变量与其他变量之间关系不大.②各地居民感知和旅游支持度存在差异,南门汤口各维度总体得分最高,北门甘棠-耿城次之,东门谭家桥第三,西门焦村得分最低.③ 门户社区与黄山风景区的空间关系、区位条件与交通格局、旅游经济发展水平等因素造成了各门户社区旅游支持度的差异.