Search for: All records

The widespread availability of geotagged data combined with modern map services allows for the accurate attachment of data to spatial networks. Applying statistical analysis, such as hotspot detection, over spatial networks is very important for precise quantification and patterns analysis, which empowers effective decision-making in various important applications. Existing hotspot detection algorithms on spatial networks either lack sufficient statistical evidence on detected hotspots, such as clustering, or they provide statistical evidence at a prohibitive computational overhead. In this paper, we propose efficient algorithms for detecting hotspots based on the network local K-function for predefined and unknown hotspot radii. The K-function is a widely adopted statistical approach for network pattern analysis that enables the understanding of the density and distribution of activities and events happening within the spatial network. However, its practical application has been limited due to the inefficiency of state-of-the-art algorithms, particularly for large-sized networks. Extensive experimental evaluation using real and synthetic datasets shows that our algorithms are up to 28 times faster than the state-of-the-art algorithms in computing hotspots with a predefined radius and up to more than four orders of magnitude faster in identifying hotspots without a predefined radius. Additionally, to address dynamic changes in the spatial network, we propose an incremental hotspot detection approach that efficiently updates hotspot computations by leveraging prior results as new events are added. 

Urban dynamics is complex and interconnected across various social and environmental systems. To better understand such dynamics, this study proposes a scalable and flexible video machine learning framework for spatiotemporal analysis of urban dynamics. The framework is based on a space–time cube representation and decomposes the cube structure along the temporal dimension into a sequence of time‐series spatial aggregation, similar to a video. State‐of‐the‐art video machine learning models including ConvLSTM, predRNN, predRNN‐V2, and E3D‐LSTM are utilized for spatiotemporal modeling and prediction. The scalability of this cyberGIS‐enabled framework is shown by its applicability to diverse geographic regions, its ability to address various urban problems, and its capacity to integrate heterogeneous geospatial data. Moreover, the framework's flexibility is further enhanced by adjustable spatial and temporal granularity. The framework's effectiveness is validated through two case studies: (1) a real‐world urban heat analysis in Cook County, Illinois, USA in 2018, which achieved an RMSE of 0.60535°C, representing a 46% improvement over established benchmarks; and (2) a simulated dataset analysis demonstrating the framework's adaptability for spatial heterogeneity and temporal changes. A series of evaluations demonstrate the effectiveness of the proposed framework in spatiotemporal analysis of complex urban dynamics. 

Place‐based spatial accessibility quantifies the distribution of access to goods and services across space. The Two‐Step Floating Catchment Area (2SFCA) family of methods have become a default tool for spatial accessibility analysis in part due to their intuitive approach and interpretability. This family of methods relies on calculating catchment areas around supply locations to estimate the area and population that may utilize them. However, these “catchment areas” are generally defined by origin‐destination matrices of travel‐time, giving us point‐to‐point distances and not polygons with actual area. This means that population geographies (census tracts, blocks, etc.) are binarily included or excluded, with no room for partial inclusion. When using nongranular data, which is often the case due to data privacy restrictions, this has the potential to cause significant errors in accessibility measurements. In this article, we propose Areal 2SFCA: a new approach that considers the area of overlap between travel‐time polygons and population geographies. We demonstrate the effectiveness of the Areal 2SFCA method using a case study that compares the Enhanced Two‐Step Floating Catchment Area (E2SFCA) and Areal E2SFCA for the state of Illinois in the USA using multiple population granularities. 

The Zika virus epidemic of 2015–16, which caused over 1 million confirmed or suspected human cases in the Caribbean and Latin America, was driven by a combination of movement of infected humans and availability of suitable habitat for mosquito species that are key disease vectors. Both human mobility and mosquito vector abundances vary seasonally, and the goal of our research was to analyze the interacting effects of disease vector densities and human movement across metapopulations on disease transmission intensity and the probability of super-spreader events. Our research uses the novel approach of combining geographical modeling of mosquito presence with network modeling of human mobility to offer a comprehensive simulation environment for Zika virus epidemics that considers a substantial number of spatial and temporal factors compared to the literature. Specifically, we tested the hypotheses that 1) regions with the highest probability of mosquito presence will have more super-spreader events during dry months, when mosquitoes are predicted to be more abundant, 2) regions reliant on tourism industries will have more super-spreader events during wet months, when they are more likely to contribute to network-level pathogen spread due to increased travel. We used the case study of Colombia, a country with a population of about 50 million people, with an annual calendar that can be partitioned into overlapping cycles of wet and dry seasons and peak tourism and off tourism seasons that drive distinct cyclical patterns of mosquito abundance and human movement. Our results show that whether the first infected human was introduced to the network during the wet versus dry season and during the tourism versus off tourism season profoundly affects the severity and trajectory of the epidemic. For example, Zika virus was first detected in Colombia in October of 2015. Had it originated in January, a dry season month with high rates of tourism, it likely could have infected up to 60% more individuals and up to 40% more super-spreader events may have occurred. In addition, popular tourism destinations such as Barranquilla and Cartagena have the highest risk of super-spreader events during the winter, whereas densely populated areas such as Medellín and Bogotá are at higher risk of sustained transmission during dry months in the summer. Our research demonstrates that public health planning and response to vector-borne disease outbreaks requires a thorough understanding of how vector and host patterns vary due to seasonality in environmental conditions and human mobility dynamics. This research also has strong implications for tourism policy and the potential response strategies in case of an emergent epidemic. 

This paper reviews trends in GeoAI research and discusses cutting-edge advances in GeoAI and its roles in accelerating environmental and social sciences. It addresses ongoing attempts to improve the predictability of GeoAI models and recent research aimed at increasing model explainability and reproducibility to ensure trustworthy geospatial findings. The paper also provides reflections on the importance of defining the science of GeoAI in terms of its fundamental principles, theories, and methods to ensure scientific rigor, social responsibility, and lasting impacts. 

Hydrological streamline delineation is critical for effective environmental management, influencing agriculture sustainability, river dynamics, watershed planning, and more. This study develops a novel approach to combining transfer learning with convolutional neural networks that capitalize on image-based pre-trained models to improve the accuracy and transferability of streamline delineation. We evaluate the performance of eleven image-based pre-trained models and a baseline model using datasets from Rowan County, North Carolina, and Covington River, Virginia in the USA. Our results demonstrate that when models are adapted to a new area, the fine-tuned ImageNet pre-trained model exhibits superior predictive accuracy, markedly higher than the models trained from scratch or those only fine-tuned on the same area. Moreover, the pre-trained model achieves better smoothness and connectivity between classified streamline channels. These findings underline the effectiveness of transfer learning in enhancing the delineation of hydrological streamlines across varied geographies, offering a scalable solution for accurate and efficient environmental modelling. 

As a decisive part in the success of Mobility-as-a-Service (MaaS), spatio-temporal dynamics modeling on mobility networks is a challenging task particularly considering scenarios where open-world events drive mobility behavior deviated from the routines. While tremendous progress has been made to model high-level spatio-temporal regularities with deep learning, most, if not all of the existing methods are neither aware of the dynamic interactions among multiple transport modes on mobility networks, nor adaptive to unprecedented volatility brought by potential open-world events. In this paper, we are therefore motivated to improve the canonical spatio-temporal network (ST-Net) from two perspectives: (1) design a heterogeneous mobility information network (HMIN) to explicitly represent intermodality in multimodal mobility; (2) propose a memory-augmented dynamic filter generator (MDFG) to generate sequence-specific parameters in an on-the-fly fashion for various scenarios. The enhanced event-aware spatio-temporal network, namely EAST-Net, is evaluated on several real-world datasets with a wide variety and coverage of open-world events. Both quantitative and qualitative experimental results verify the superiority of our approach compared with the state-of-the-art baselines. What is more, experiments show generalization ability of EAST-Net to perform zero-shot inference over different open-world events that have not been seen.