Drive-by sensing, using vehicles as mobile sensors to collect environmental data, offers high spatiotemporal resolution monitoring. In drive-by sensing tasks, urban areas with intense human activity or pollution – termed high-priority hotspots – demand frequent sensing for adequate data collection. However, prior studies rarely address optimizing vehicle selection to enhance hotspot coverage frequency while maintaining non-hotspot coverage. This study formalized this problem as the Maximal Hotspot Regular Coverage Problem and proposed an Adaptive Level-Aware Vehicle Selection algorithm. Using air pollution hotspot sensing in Beijing as an empirical case, results show the advantage of the proposed algorithm over other baselines: Random selection (RS), Non-hotspot Greedy Adding and Multi-type hotspot Greedy MaxMin, covering 29.99% of hotspots and 77.08% non-hotspot zones with 1000 sensors on average. The spatial distribution of covered area and statistical distribution of average visits per hour reveals a large spatial range and high revisit times of the method, and its advantage over baselines in a hybrid sensing scenario is also proven (with 54.18% daily average coverage among all zones and 31.22% on hotspots). This study provides a method to advance the coverage of hotspots in single-type and hybrid sensing scenarios, inspiring a future extension of optimization based on the proposed algorithm.

The development of geospatial artificial intelligence (GeoAI) systems depends on the ability to learn effective representations of places. To learn accurate place representations from spatial interactions, it is important to extract features that capture both the spatial and non-spatial driving factors. However, existing methods lack a robust interpretation and the explanatory power of the learned representations on spatial factors remains unexplored. Here, we propose an approach to learning place representations from spatial interactions. Our method is inspired by flow allocation, which is the main focus of single-constrained gravity models. We first validate the method on synthetic flows with known driving factors and then apply it to multi-scale real-world flows. Results show that the learned representations can effectively capture features that explain place characteristics, along with the impact of spatial impedance. Our study not only contributes an efficient method to learn place representations from spatial interactions but also offers insights into pre-training procedures in GeoAI.

Existing regionalization methods tend to be either spatial attribute- or spatial interaction-based, while real-world tasks usually involve both considerations to satisfy multiple objectives simultaneously. In this research, we propose Multilayer Network Community Detection and Kernel Extension (MNCD-KE), a two-step regionalization framework, as a feasible solution for such tasks. First, spatial attributes are embedded into attributes of nodes in a spatial interaction-defined multilayer network, and the kernel and marginal parts of the regions are determined by giving the membership value of the regionalization units to network communities. Second, the final result is obtained through a kernel extension process considering geographical constraints, including spatial contiguity, size balance, morphological regularity, and existing boundary consistency of the regions. Empirical experiments show that the proposed method yields outcomes that, in maintaining comparable performances with most baseline algorithms with either ‘attribute’ or ‘interaction’ objectives as measured by the respective criteria, simultaneously meet the dual objectives with results intuitively comprehensible. Its low computing costs and parameter adjustment flexibility make the proposed framework a convenient approach for real-world multi-objective regionalization tasks. We conclude the research with discussions on the boundary conditions for the framework to work and their relevance to city science theories, along with practical implications.

In spatial regression models, spatial heterogeneity may be considered with either continuous or discrete specifications. The latter is related to delineation of spatially connected regions with homogeneous relationships between variables (spatial regimes). Although various regionalization algorithms have been proposed and studied in the field of spatial analytics, methods to optimize spatial regimes have been largely unexplored. In this paper, we propose two new algorithms for spatial regime delineation, two-stage K-Models and Regional-K-Models. We also extend the classic Automatic Zoning Procedure to a spatial regression context. The proposed algorithms are applied to a series of synthetic datasets and two real-world datasets. Results indicate that all three algorithms achieve superior or comparable performance to existing approaches, while the two-stage K-Models algorithm largely outperforms existing approaches on model fitting, region reconstruction and coefficient estimation. Our work enriches the spatial analytics toolbox to explore spatial heterogeneous processes.

About 30 years have passed since Michael F. Goodchild proposed the term geographical information science (GIScience) in 1992. In the past 30 years, GIScience has made great progress in expanding research findings and perfecting theories and methods. To understand the development progress of GIScience, this research conducts a bibliometric analysis of 9400 publications between 1991 and 2020 in 10 international refereed journals and 2 international conferences of GIScience. We analyze the publication statistics and trends in GIScience from two aspects of journals/conferences and countries/territories. Based on the community detection of the citation network, we extract 15 research themes and show their leading authors and highly cited articles. Furthermore, the change of publication number in different themes over time can indicate the evolution of some research focuses in GIScience. The results demonstrate that the publication proportions of some themes grow rapidly, such as “moving object,” “volunteered geographic information,” and “geographically weight regression,” while the publication proportions of some themes are decreasing, such as “digital elevation model,” “planning support system,” and “ontology.” In the discussion, the journal distribution of papers on different themes is discussed. Moreover, we suggest a few research directions that are worthy of attention in the future.

Cross‐city point of interest (POI) recommendation for tourists in an unfamiliar city has high application value but is challenging due to the data sparsity. Most existing models attempt to alleviate the sparsity problem by learning the user preference transfer and drift. However, they either fail to simultaneously model the preference transfer and drift in both long‐ and short‐term user preferences or cannot accomplish the task of the next POI recommendation, which is crucial for a wide spectrum of applications ranging from transportation and urban planning to advertising. To address the limitation, we proposed a user preference transfer and drift network (UPTDNet) for cross‐city next POI recommendation. UPTDNet excels at cross‐city recommendations by learning the transfer and drift of both long‐ and short‐term preferences. For short‐term preference, dual recurrent neural network‐based (RNN‐based) branches are designed to model preference transfer from tourist’s current city and drift among different user roles. For long‐term preference, a mapping function and user similarity calculation are employed for preference transfer from the tourist’s home city and drift among individual users. Experiments are conducted on the Gowalla and Foursquare datasets, and the results show that UPTDNet consistently and significantly outperforms state‐of‐the‐art models by an average of 10.22% to 22.63% in the next POI recommendation task. Ablation study and further analysis validate the effectiveness and plausibility of considering both user preference transfer and drift in the cross‐city recommendation.

Simulating urban morphology with location attributes is a challenging task in urban science. Recent studies have shown that Generative Adversarial Networks (GANs) have the potential to shed light on this task. However, existing GAN-based models are limited by the sparsity of urban data and instability in model training, hampering their applications. Here, we propose a GAN framework with geographical knowledge, namely Metropolitan GAN (MetroGAN), for urban morphology simulation. We incorporate a progressive growing structure to learn hierarchical features and design a geographical loss to impose the constraints of water areas. Besides, we propose a comprehensive evaluation framework for the complex structure of urban systems. Results show that MetroGAN outperforms the state-of-the-art urban simulation methods by over 20% in all metrics. Inspiringly, using physical geography features singly, MetroGAN can still generate shapes of the cities. These results demonstrate that MetroGAN solves the instability problem of previous urban simulation GANs and is generalizable to deal with various urban attributes.

As the essence of urban spatiotemporal interaction systems, hubs and centers empower cities to enhance socioeconomic prosperity and sustainability. However, a city manifests a time-evolving spatial interaction network with latent temporal interactions and irregular spatial partitions. This phenomenon is termed the spatiotemporal inconsistency problem. The aggregate, single-layer network model is defective for capturing the importance of locations in such time-evolving spatial interaction systems. This article therefore proposes a novel multilayer network model based on the nature of inherent spatial and temporal dependencies of urban interactions. First, the spatial agglomeration and the temporal correlation are explicitly modeled in multilayer networks for alleviating the spatiotemporal inconsistency problem. Secondly, generalized centrality metrics from a single-layered static network to the multi-layered dynamic network are acquired in order to discover grouped hub locations over time. Lastly, the capability of the proposed method is evaluated by an empirical analysis of the taxi mobility networks of Beijing, China, from 2012 to 2017. The empirical analysis indicates that the proposed method enables the identification of typical hub locations clustered in space and stable over time. This ability is essential to understand the centrality of locations informed by noisy and inconsistent data in their spatial and temporal dimensions.

Location prediction based on spatio-temporal footprints in social media is instrumental to various applications, such as travel behavior studies, crowd detection, traffic control, and location-based service recommendation. In this study, we propose a model that uses geotags of social media to predict the potential area containing users’ next locations. In the model, we utilize HiSpatialCluster algorithm to identify clustering areas (CAs) from check-in points. CA is the basic spatial unit for predicting the potential area containing users’ next locations. Then, we use the LINE (Large-scale Information Network Embedding) to obtain the representation vector of each CA. Finally, we apply BiLSTM-CNN (Bidirectional Long Short-Term Memory-Convolutional Neural Network) for location prediction. The results show that the proposed ensemble model outperforms the single LSTM or CNN model. In the case study that identifies 100 CAs out of Weibo check-ins collected in Wuhan, China, the Top-5 predicted areas containing next locations amount to an 80% accuracy. The high accuracy is of great value for recommendation and prediction on areal unit.

Big geo-data are often aggregated according to spatio-temporal units for analyzing human activities and urban environments. Many applications categorize such data into groups and compare the characteristics across groups. The intergroup differences vary with spatio-temporal units, and the essential is to identify the spatio-temporal units with apparently different data characteristics. However, spatio-temporal dependence, data variety, and the complexity of tasks impede an effective unit assessment. Inspired by the applications to extract critical image components based on explainable artificial intelligence (XAI), we propose a spatio-temporal layer-wise relevance propagation method to assess spatio-temporal units as a general solution. The method organizes input data into an extensible three-dimensional tensor form. We provide two means of labeling the spatio-temporal tensor data for typical geographical applications, using temporally or spatially relevant information. Neural network training proceeds to extract the global and local characteristics of data for corresponding analytical tasks. Then the method propagates classification results backward into units as obtained task-specific importance. A case study with taxi trajectory data in Beijing validates the method. The results prove that the proposed method can evaluate the task-specific importance of spatio-temporal units with dependence. This study also attempts to discover task-related knowledge using XAI.

Spatial interpolation is a traditional geostatistical operation that aims at predicting the attribute values of unobserved locations given a sample of data defined on point supports. However, the continuity and heterogeneity underlying spatial data are too complex to be approximated by classic statistical models. Deep learning models, especially the idea of conditional generative adversarial networks (CGANs), provide us with a perspective for formalizing spatial interpolation as a conditional generative task. In this article, we design a novel deep learning architecture named conditional encoder-decoder generative adversarial neural networks (CEDGANs) for spatial interpolation, therein combining the encoder-decoder structure with adversarial learning to capture deep representations of sampled spatial data and their interactions with local structural patterns. A case study on elevations in China demonstrates the ability of our model to achieve outstanding interpolation results compared to benchmark methods. Further experiments uncover the learned spatial knowledge in the model’s hidden layers and test the potential to generalize our adversarial interpolation idea across domains. This work is an endeavor to investigate deep spatial knowledge using artificial intelligence. The proposed model can benefit practical scenarios and enlighten future research in various geographical applications related to spatial prediction.

The spatial hierarchy of part-whole relationships is an essential characteristic of the platial world. Constructing spatial hierarchies of places is valuable in association analysis and qualitative spatial reasoning. The emergence of large amounts of geotagged user-generated content provides strong support for modelling places. However, the vague nature of places and the complex spatial relationships among places make it intractable to understand and represent the hierarchies among places. In this paper, we introduce a fuzzy formal concept analysis-based approach to uncovering the spatial hierarchies among vague places. Each place is represented as a concept that consists of its extent and its intent. Based on the place concepts, the spatial hierarchies are generated and expressed as a graph that is easy to comprehend and contains abundant information on spatial relations. We also demonstrate the rationality of our result by comparing it with the result of a questionnaire survey.

Spatial interactions underlying consecutive sequential snapshots of spatial distributions, such as the migration flows underlying temporal population snapshots, can reflect the details of spatial evolution processes. In the era of big data, we have access to individual-level data, but the acquisition of high-quality spatial interaction data remains a challenging problem. Most research has been focused on distributions of movable objects or the modelling of spatial interaction patterns, with few attempts to identify hidden spatial interaction patterns from temporal transitions of spatial distributions. In this article, we introduced an approach to infer spatial interaction patterns from sequential snapshots of spatial population distributions by incorporating linear programming and the spatial constraints of human movement. Experiments using synthetic data were conducted using four simple scenarios to explore the characteristics of our method. The proposed method was used to extract interurban flows of migrants during the Chinese Spring Festival in 2016. Our research demonstrated the feasibility of using discrete multi-temporal snapshots of population distributions in space to infer spatial interaction patterns and offered a general analytical framework from snapshot data to spatial interaction patterns.

Information about the functions of urban buildings is helpful not only for developing a better understanding of how cities work, but also for establishing a basis for policy makers to evaluate and improve the effectiveness of urban planning. Despite these advantages, however, and perhaps simply due to a lack of available data, few academic studies to date have succeeded in integrating multi-source ‘big data’ to examine urban land use at the building level. Responding to this deficiency, this study integrated multi-source big data (WeChat users’ real-time location records, taxi GPS trajectories data, Points of Interest (POI) data, and building footprint data from high-resolution Quickbird images), and applied the proposed density-based method to infer the functions of urban buildings in Tianhe District, Guangzhou, China. The results of the study conformed to an overall detection rate of 72.22%. When results were verified against ground-truth investigation data, the accuracy rate remained above 65%. Two important conclusions can be drawn from our analysis: 1.The use of WeChat data delivers better inference results than those obtained using taxi data when used to identify residential buildings, offices, and urban villages. Conversely, shopping centers, hotels, and hospitals, were more easily identified using taxi data. 2. The use of integrated multi-source big data is more effective than single-source big data in revealing the relation between human dynamics and urban complexes at the building scale.

Land use classification has benefited from the emerging big data, such as mobile phone records and taxi trajectories. Temporal activity variations derived from these data have been used to interpret and understand the land use of parcels from the perspective of social functions, complementing the outcome of traditional remote sensing methods. However, spatial interaction patterns between parcels, which could depict land uses from a perspective of connections, have rarely been examined and analysed. To leverage spatial interaction information contained in the above-mentioned massive data sets, we propose a novel unsupervised land use classification method with a new type of place signature. Based on the observation that spatial interaction patterns between places of two specific land uses are similar, the new place signature improves land use classification by trading off between aggregated temporal activity variations and detailed spatial interactions among places. The method is validated with a case study using taxi trip data from Shanghai.

This article provides a novel and practical approach for investigating the characteristics of intercity telecommunication network whose overall and complete information is unavailable. Using a mobile phone call data set covering 4.39 million subscribers registered in a particular region, we construct two intercity mobile communication subnets and infer characteristics of the whole intercity mobile communication network of China. Results confirm that intercity communication intensity is characterized by the gravity model. The communication intensity based on mobile call number decreases along the distance with a scaling exponent 0.5, whereas the scaling exponent for the communication intensity based on mobile call duration is 0.4. Moreover, we uncover the rank-size distribution of tie strength (mobile call number and duration) between a city and its neighbours. The rank-size law of tie strengths between cities is mainly determined by the rank-size distribution of cities. The distance between cities plays a less decisive role than the size distribution in the network, but significantly impacts mobile communication patterns. The call duration of individual intercity mobile communication is generally positively correlated to the communication distance, explaining why the distance decay of communication intensity based on call durations is slower than that based on call numbers. The contribution of this research is twofold. First, we identify the distance decay effect in intercity mobile communications of China and uncover the dominant impact of the rank-size distribution of cities. Second, a method for estimating the properties of the whole network according to the observed interactions of its subnets is developed.

One common problem with geographic data is that, for a specific geographic event, only occurrence information is available; information about the absence of the event is not available. We refer to these specific types of geospatial data as geographic one-class data (GOCD). Predicting the potential spatial distributions that a particular geographic event may occur from GOCD is difficult because traditional binary classification methods that require availability of both positive and negative training samples cannot be used. The objective of this research is to define GOCD and propose novel approaches for modelling potential spatial distributions of geographic events using GOCD. We investigate the effectiveness of one-class support vector machine (OCSVM), maximum entropy (MAXENT) and the newly proposed positive and unlabelled learning (PUL) algorithm for solving GOCD problems using a case study: species distribution modelling from synthetic data. Our experimental results indicate that generally OCSVM, MAXENT and PUL are effective in modelling the GOCD. Each method has advantages and disadvantages, but PUL seems to be the most promising method.

We describe a method of morphometric characterisation of landform from digital elevation models (DEMs). The method is implemented first by classifying every location into morphometric classes based on the mathematical shape of a locally fitted quadratic surface and its positional relationship with the analysis window. Single‐scale fuzzy terrain indices of peakness, pitness, passness, ridgeness, and valleyness are then calculated based on the distance of the analysis location from the ideal cases. These can then be combined into multi‐scale terrain indices to summarise terrain information across different operational scales. The algorithm has four characteristics: (1) the ideal cases of different geomorphometric features are simply and clearly defined; (2) the output is spatially continuous to reflect the inherent fuzziness of geomorphometric features; (3) the output is easily combined into a multi‐scale index across a range of operational scales; and (4) the standard general morphometric parameters are quantified as the first and second order derivatives of the quadratic surface. An additional benefit of the quadratic surface is the derivation of the R 2 goodness of fit statistic, which allows an assessment of both the reliability of the results and the complexity of the terrain. An application of the method using a test DEM indicates that the single‐ and multi‐scale terrain indices perform well when characterising the different geomorphometric features.

Locality information for specimens of geological, biological, and cultural objects is traditionally stored as textual descriptions. With an increasing demand for natural and cultural information, the lack of spatially explicit descriptions has become a major barrier to the management and analysis of these data using geographic information systems. In this paper, we propose a method to georeference descriptive data, using an uncertainty field model to represent the distribution of a locality based on two types of uncertainties: uncertainty of reference objects, and the uncertainty of spatial relationships. We propose probability distributions for each known form of these two types of uncertainties and present a probabilistic method to georeference localities based on the integration of different uncertainty sources.

Geospatial data modelling is dominated by the distinction between continuous‐field and discrete‐object conceptualizations. However, the boundary between them is not always clear, and the field view is more fundamental in some respects than the object view. By viewing a set of objects as an object field and unifying it with conventional field models, a new concept, the General Field (G‐Field) model, is proposed. In this paper, the properties of G‐Field models, including domain, range, and categorization, are discussed. As a summary, a descriptive framework for G‐Field models is proposed. Then, some common geospatial operations in geographic information systems are reconsidered from the G‐Field perspective. The geospatial operations are classified into order‐increasing operations and non‐order‐increasing operations, depending on changes induced in the G‐Field's order. Generally, the order can be viewed as an indicator of the level of information extraction of geospatial data. It is thus possible to integrate the concept of order with a geo‐workflow management system to support geographic semantics.