Accurately attributing user visits to specific Points of Interest (POIs) is a foundational task for mobility analytics, personalized services, marketing and urban planning. However, POI attribution remains challenging due to GPS inaccuracies, typically ranging from 2 to 20 meters in real-world settings, and the high spatial density of POIs in urban environments, where multiple venues can coexist within a small radius (e.g., over 50 POIs within a 100-meter radius in dense city centers). Relying on proximity is therefore often insufficient for determining which POI was actually visited. We introduce POIFormer, a novel Transformer-based framework for accurate and efficient POI attribution. Unlike prior approaches that rely on limited spatiotemporal, contextual, or behavioral features, POIFormer jointly models a rich set of signals, including spatial proximity, visit timing and duration, contextual features from POI semantics, and behavioral features from user mobility and aggregated crowd behavior patterns-using the Transformer's self-attention mechanism to jointly model complex interactions across these dimensions. By leveraging the Transformer to model a user's past and future visits (with the current visit masked) and incorporating crowd-level behavioral patterns through pre-computed kernel density estimates (KDEs), POIFormer enables accurate, efficient attribution in large, noisy mobility datasets. Its architecture supports generalization across diverse data sources and geographic contexts while avoiding reliance on hard-to-access or unavailable data layers, making it practical for real-world deployment. Extensive experiments on real-world mobility datasets demonstrate significant improvements over existing baselines, particularly in challenging real-world settings characterized by spatial noise and dense POI clustering.
Recent advances in large foundation models (FMs) have enabled learning general-purpose representations in natural language, vision, and audio. Yet geospatial artificial intelligence (GeoAI) still lacks widely adopted foundation models that generalize across tasks that require joint reasoning over geospatial objects and human mobility. Such tasks are crucial as mobility, along with satellite imagery, street view, and text, is a core modality for understanding the physical world. We argue that a key bottleneck is the absence of unified, general-purpose, and transferable representations for geospatially embedded objects (GEOs). Such objects include points, polylines, and polygons in geographic space, enriched with semantic context and critical for geospatial reasoning. Much current GeoAI research compares GEOs to tokens in language models, where patterns of human movement and spatiotemporal interactions yield contextual meaning similar to patterns of words in text. However, modeling GEOs introduces challenges fundamentally different from language, including spatial continuity, variable scale and resolution, temporal dynamics, and data sparsity. Moreover, privacy constraints and global variation in mobility further complicates modeling and generalization. This paper formalizes these challenges, identifies key representational gaps, and outlines research directions for building foundation models that learn behavior-informed, transferable representations of GEOs from large-scale human mobility data, as well as static contextual information such as points of interest, object shapes and spatio-temporal semantics.
Inferring social relationships from mobility data is crucial for many applications because it reflects real-world connections among people. However, large-scale trajectory datasets with ground-truth social ties are exceedingly scarce, making it difficult to train deep models for relationship inference. To address this gap, we propose a transferable social relationship inference framework that can be trained on one high-quality, labeled dataset and then generalized to new datasets, even from different cities. Our framework rests on the key insight that social bonds depend largely on the frequency of individual meetings and the popularity of those meeting locations, both of which can be inferred statistically from raw trajectory data, irrespective of the underlying geographic semantics. It comprises two main modules: 1) Universal Social Relationship Classifier (USRC): A model trained to infer social relationships from trajectory data, and 2) Spatial Embedding Transfer (SET): A location embedding alignment technique that adapts new datasets to the pre-trained USRC model. By aligning location embeddings, SET module enables the pre-trained USRC to interpret previously unseen datasets without extra supervision. Experiments on five public datasets demonstrate that our method achieves state-of-the-art performance in zero-shot social relationship inference, surpassing other unsupervised, and in some cases, even supervised, approaches. Additionally, the SET module significantly improves location embedding alignment, outperforming existing baseline methods. The source code and data are available at https://github.com/chuchen2017/SET.
Abnormal human mobility patterns often signal disruptions, emergencies, or health-related risks, making their detection critical for applications in public safety, urban monitoring, and healthcare. Existing approaches for human mobility anomaly detection typically focus on either identifying visits to unusual places or overall deviations from individual- and population-level norms at the agent-level. However, these methods often (1) overlook fine-grained temporal anomalies, and (2) lack interpretability, as they do not reveal which specific spatiotemporal components of a visit contribute to its anomalous nature. To overcome these limitations, we present ICAD (Interpretable Component-wise Anomaly Detection), a self-supervised autoregressive model that detects both spatial and temporal anomalies by modeling deviations in an individual's visit-level mobility behavior. ICAD is trained on normal visit sequences using a next-visit prediction objective to learn the distribution of visits under regular conditions. At inference, it computes component-wise anomaly scores for each visit by measuring relative divergence from the learned distribution of normal behavior. Specifically, ICAD proposes a top-k deviation metric for discrete spatial anomalies and introduces a novel relative mode-based scoring function for detecting temporal anomalies in continuous time. Experiments on a large scale synthetic human mobility dataset show that ICAD outperforms prior methods in both visit-level and agent-level anomaly detection. For reproducability purposes, the source code is accessible at https://github.com/USC-InfoLab/ICAD.
In this study, we present a convolutional neural network (CNN) architecture, GeoConv, designed to improve the accuracy and adaptability of deep learning models using satellite imagery. Traditional CNNs, such as ResNet18, employ fixed-weight convolutional layers - i.e., layers that leverage the same set of weights for each input observation. However, these models can struggle to capture context-specific features inherent in satellite images, which may vary significantly across different geographic regions. To address this challenge, the GeoConv model utilizes dynamic weights that adapt based on the input image coordinates, allowing the model to tailor its feature extraction process to the unique characteristics of different geographic regions. Through experiments, we illustrate the utility of this approach in a case study which leverages satellite imagery to estimate household wealth across 11 countries, with GeoConv explaining an additional 10.12% of the variance in the data compared to a ResNet18 model. These results underscore the importance of incorporating spatially adaptive mechanisms in handling the variability present in satellite imagery. Code is available at: https://github.com/heatherbaier/geoconv
Accessing realistic human movements (aka trajectories) is essential for many application domains, such as urban planning, transportation, and public health. However, due to privacy and commercial concerns, real-world trajectories are not readily available, giving rise to an important research area of generating synthetic but realistic trajectories. Inspired by the success of deep neural networks (DNN), data-driven methods learn the underlying human decision-making mechanisms and generate synthetic trajectories by directly fitting real-world data. However, these DNN-based approaches do not exploit people's moving behaviors (e.g., work commute, shopping purpose), significantly influencing human decisions during the generation process. This paper proposes MBP-GAIL, a novel framework based on generative adversarial imitation learning that synthesizes realistic trajectories that preserve moving behavior patterns in real data. MBP-GAIL models temporal dependencies by Recurrent Neural Networks (RNN) and combines the stochastic constraints from moving behavior patterns and spatial constraints in the learning process. Through comprehensive experiments, we demonstrate that MBP-GAIL outperforms state-of-the-art methods and can better support decision making in trajectory simulations.
Measuring spatial accessibility to healthcare resources and facilities has long been an important problem in public health. For example, during disease outbreaks, sharing spatial accessibility data such as individual travel distances to health facilities is vital to policy making and designing effective interventions. However, sharing these data may raise privacy concerns, as information about individual data contributors (e.g., health status and residential address) may be disclosed. In this work, we investigate those unintended information leakage in spatial accessibility analysis. Specifically, we are interested in understanding whether sharing data for spatial accessibility computations may disclose individual participation (i.e., membership inference) and personal identifiable information (i.e., address inference). Furthermore, we propose two provably private algorithms that mitigate those privacy risks. The evaluation is conducted with real population and healthcare facilities data from Mecklenburg county, NC and Nashville, TN. Compared to state-of-the-art privacy practices, our methods effectively reduce the risks of membership and address disclosure, while providing useful data for spatial accessibility analysis.
Contact tracing is gaining its importance in controlling the spread of COVID-19. However, the enormous volume of the frequently sampled tracing data brings major challenges for real-time processing. In this paper, we propose a GPU-based real-time contact tracing system based on spatial proximity queries with temporal constraints using location data. We provide dynamic indexing of moving objects using an adaptive partitioning schema on GPU with extremely low overhead. Our system optimizes the retrieval of contacted pairs to match both the requirements of contact tracing scenarios and GPU centered parallelism. We propose an efficient contacts evaluation mechanism to keep only the spatially and temporally valid contacts. Our experiments demonstrate that the system can achieve sub-second level response for large-scale contact tracing of tens of millions of people, with two magnitudes of performance boost over CPU based approach.
Much effort has been devoted to support high performance spatial queries on large volumes of spatial data in distributed spatial computing systems, especially in the MapReduce paradigm. Recent works have focused on extending spatial MapReduce frameworks to leverage high performance in-memory distributed processing capabilities of systems such as Spark. However, the performance advantage comes with the requirement of having enough memory and comprehensive configuration. Failing to fulfill this falls back to disk IO, defeating the purpose of such systems or in worst case gets out of memory and fails the job. The problem is aggravated further for spatial processing since the underlying in-memory systems are oblivious of spatial data features and characteristics. In this paper we present SparkGIS - an in-memory oriented spatial data querying system for high throughput and low latency spatial query handling by adapting Apache Spark's distributed processing capabilities. It supports basic spatial queries including containment, spatial join and k-nearest neighbor and allows extending these to complex query pipelines. SparkGIS mitigates skew in distributed processing by supporting several dynamic partitioning algorithms suitable for a rich set of contemporary application scenarios. Multi-level global and local, pre-generated and on-demand in-memory indexes, allow SparkGIS to prune input data and apply compute intensive operations on a subset of relevant spatial objects only. Finally, SparkGIS employs dynamic query rewriting to gracefully manage large spatial query workflows that exceed available distributed resources. Our comparative evaluation has shown that the performance of SparkGIS is on par with contemporary Spark based platforms for relatively smaller queries and outperforms them for larger data and memory intensive workflows by dynamic query rewriting and efficient spatial data management.
Air quality models are important for studying the impact of air pollutant on health conditions at a fine spatiotemporal scale. Existing work typically relies on area-specific, expert-selected attributes of pollution emissions (e,g., transportation) and dispersion (e.g., meteorology) for building the model for each combination of study areas, pollutant types, and spatiotemporal scales. In this paper, we present a data mining approach that utilizes publicly available OpenStreetMap (OSM) data to automatically generate an air quality model for the concentrations of fine particulate matter less than 2.5 μm in aerodynamic diameter at various temporal scales. Our experiment shows that our (domain-) expert-free model could generate accurate PM2.5 concentration predictions, which can be used to improve air quality models that traditionally rely on expert-selected input. Our approach also quantifies the impact on air quality from a variety of geographic features (i.e., how various types of geographic features such as parking lots and commercial buildings affect air quality and from what distance) representing mobile, stationary and area natural and anthropogenic air pollution sources. This approach is particularly important for enabling the construction of context-specific spatiotemporal models of air pollution, allowing investigations of the impact of air pollution exposures on sensitive populations such as children with asthma at scale.
Recent advances in digital pathology make it possible to support 3D tissue-based investigation of human diseases at extremely high resolutions. Exploring spatial relationships and patterns among massive 3D micro-anatomic biological objects such as blood vessels and cells derived from 3D pathology image volumes plays a critical role in studying human diseases. In this paper, we present our work on building an effective and scalable in-memory based spatial query system iSPEED for large-scale 3D data with complex structures. To achieve low latency, iSPEED stores data in memory with effective progressive compression for each 3D object with successive levels of detail. To minimize search space and computation cost, iSPEED pregenerates global spatial indexes in memory and employs on-demand indexing at run-time. In particular, iSPEED exploits structural indexing for complex structured objects in distance based queries. iSPEED provides a 3D spatial query engine that can be invoked on-demand to run many instances in parallel implemented with, but not limited to, MapReduce. iSPEED builds in-memory indexes and decompresses data on-demand, which has minimal memory footprint. We evaluate iSPEED with two representative queries: 3D spatial joins and 3D spatial proximity estimation. Our experiments demonstrate that iSPEED significantly improves the performance over traditional non-memory based spatial query systems.
3D analytical pathology imaging examines high resolution 3D image volumes of human tissues to facilitate biomedical research and provide potential effective diagnostic assistance. Such approach - quantitative analysis of large- scale 3D pathology image volumes - generates tremendous amounts of spatially derived 3D micro-anatomic objects, such as 3D blood vessels and nuclei. Spatial exploration of such massive 3D spatial data requires effective and efficient querying methods. In this paper, we present a scalable and efficient 3D spatial query system for querying massive 3D spatial data based on MapReduce. The system provides an on-demand spatial querying engine which can be executed with as many instances as needed on MapReduce at runtime. Our system supports multiple types of spatial queries on MapReduce through 3D spatial data partitioning, customizable 3D spatial query engine, and implicit parallel spatial query execution. We utilize multi-level spatial indexing to achieve efficient query processing, including global partition indexing for data retrieval and on-demand local spatial indexing for spatial query processing. We evaluate our system with two representative queries: 3D spatial joins and 3D k-nearest neighbor query. Our experiments demonstrate that our system scales to large number of computing nodes, and efficiently handles data-intensive 3D spatial queries that are challenging in analytical pathology imaging.
Support of high performance queries on large volumes of spatial data becomes increasingly important in many application domains, including geospatial problems in numerous fields, location based services, and emerging scientific applications that are increasingly data- and compute-intensive. The emergence of massive scale spatial data is due to the proliferation of cost effective and ubiquitous positioning technologies, development of high resolution imaging technologies, and contribution from a large number of community users. There are two major challenges for managing and querying massive spatial data to support spatial queries: the explosion of spatial data, and the high computational complexity of spatial queries. In this paper, we present Hadoop-GIS - a scalable and high performance spatial data warehousing system for running large scale spatial queries on Hadoop. Hadoop-GIS supports multiple types of spatial queries on MapReduce through spatial partitioning, customizable spatial query engine RESQUE, implicit parallel spatial query execution on MapReduce, and effective methods for amending query results through handling boundary objects. Hadoop-GIS utilizes global partition indexing and customizable on demand local spatial indexing to achieve efficient query processing. Hadoop-GIS is integrated into Hive to support declarative spatial queries with an integrated architecture. Our experiments have demonstrated the high efficiency of Hadoop-GIS on query response and high scalability to run on commodity clusters. Our comparative experiments have showed that performance of Hadoop-GIS is on par with parallel SDBMS and outperforms SDBMS for compute-intensive queries. Hadoop-GIS is available as a set of library for processing spatial queries, and as an integrated software package in Hive.
Support of high performance queries on large volumes of scientific spatial data is becoming increasingly important in many applications. This growth is driven by not only geospatial problems in numerous fields, but also emerging scientific applications that are increasingly data- and compute-intensive. For example, digital pathology imaging has become an emerging field during the past decade, where examination of high resolution images of human tissue specimens enables more effective diagnosis, prediction and treatment of diseases. Systematic analysis of large-scale pathology images generates tremendous amounts of spatially derived quantifications of micro-anatomic objects, such as nuclei, blood vessels, and tissue regions. Analytical pathology imaging provides high potential to support image based computer aided diagnosis. One major requirement for this is effective querying of such enormous amount of data with fast response, which is faced with two major challenges: the "big data" challenge and the high computation complexity. In this paper, we present our work towards building a high performance spatial query system for querying massive spatial data on MapReduce. Our framework takes an on demand index building approach for processing spatial queries and a partition-merge approach for building parallel spatial query pipelines, which fits nicely with the computing model of MapReduce. We demonstrate our framework on supporting multi-way spatial joins for algorithm evaluation and nearest neighbor queries for microanatomic objects. To reduce query response time, we propose cost based query optimization to mitigate the effect of data skew. Our experiments show that the framework can efficiently support complex analytical spatial queries on MapReduce.
