Representation learning for geospatial data

In the Internet of thing (IoT), with the geographic location of geospatial sensor data and the global positioning systems, location-based services (LBSs) can provide powerful location-aware IoT applications for mobile clients according to their current locations. For LBSs, a k-nearest neighbor (kNN) search can provide a mobile client with geospatial sensor data of k-nearest spatial points of interest (POIs) according to its current location. In this paper, we propose a spatial air index with neighbor information to organize IoT geospatial sensor data for processing kNN searches in the wireless broadcast systems. Since the answered POIs may be neighbors of each other, we add neighbor information to the index structure, which is interleaved with geospatial sensor data, to speed up the query processing. To avoid unnecessary examination of geospatial sensor data from the wireless channel, the proposed method provides the centroid of geospatial data and the corresponding longest distance between the centroid and geospatial data in the region. With this information, the query processing of a kNN search can quickly determine whether to skip examining this region, saving energy consumption of the mobile device. Performance evaluations have verified that the proposed method outperforms the distributed spatial index.

Multi-level urban street representation with street-view imagery and hybrid semantic graph

Heterogeneous graph-based joint representation learning for users and POIs in location-based social network

Geospatial Big Data analytics are changing the way that businesses operate in many industries. Although a good number of research works have reported in the literature on geospatial data analytics and real-time data processing of large spatial data streams, only a few have addressed the full geospatial big data analytics project lifecycle and geospatial data science project lifecycle. Big data analysis differs from traditional data analysis primarily due to the volume, velocity and variety characteristics of the data being processed. One of a motivation of introducing new framework is to address these big data analysis challenges. Geospatial data science projects differ from most traditional data analysis projects because they could be complex and in need of advanced technologies in comparison to the traditional data analysis projects. For this reason, it is essential to have a process to govern the project and ensure that the project participants are competent enough to carry on the process. To this end, this paper presents, new geospatial big data mining and machine learning framework for geospatial data acquisition, data fusion, data storing, managing, processing, analysing, visualising and modelling and evaluation. Having a good process for data analysis and clear guidelines for comprehensive analysis is always a plus point for any data science project. It also helps to predict required time and resources early in the process to get a clear idea of the business problem to be solved.

Geospatial data exists in a variety of formats, including rasters, vector data, and large-scale geospatial databases. There exists an ever-growing number of sensors that are collecting this data, resulting in the explosive growth and scale of high-resolution remote sensing geospatial data collections. A particularly challenging domain of geospatial data processing involves mining information from high resolution remote sensing imagery. The prevalence of high-resolution raster geospatial data collections represents a significant data challenge, as a single remote sensing image is composed of hundreds of millions of pixels. We have developed a robust application framework which exploits graphics processing unit (GPU) clusters to perform high-throughput geospatial data processing. We process geospatial raster data concurrently across tiles of large geospatial data rasters, utilizing GPU co-processors driven by CPU threads to extract refined geospatial information. The framework can produce output rasters or perform image information mining to write data into a geospatial database.

Urban functional zones (UFZs) are related to people’s daily activities. Accurate recognition of UFZs is of great significance for an in-depth understanding of the complex urban system and optimizing the urban spatial structure. Emerging geospatial big data provide new ideas for humans to recognize urban functional zones. Point-of-interest (POI) data have achieved good results in the recognition of UFZs. However, since humans are the actual users of urban functions, and POI data only reflect static socioeconomic characteristics without considering the semantic and temporal features of dynamic human activities, it leads to an incomplete and insufficient representation of complex UFZs. To solve these problems, we proposed a data-synthesis-driven approach to quantify and analyze the distribution and mixing of urban functional zones. Firstly, representation learning is used to mine the spatial semantic features, activity temporal features, and activity semantic features that are embedded in POI data and social media check-in data from spatial, temporal, and semantic aspects. Secondly, a weighted Stacking ensemble model is used to fully integrate the advantages between different features and classifiers to infer the proportions of urban functions and dominant functions of each urban functional zone. A case study within the 5th Ring Road of Beijing, China, is used to evaluate the proposed method. The results show that the approach combining dynamic and static features of POI data and social media data effectively represents the semantic information of UFZs, thereby further improving the accuracy of UFZ recognition. This work can provide a reference for uncovering the hidden linkages between human activity characteristics and urban functions.

Next Point-of-Interest (POI) recommendation is an essential part of the flourishing location-based applications, where the demands of users are not only conditioned by their recent check-in behaviors but also by the critical influence stemming from geographical dependencies among POIs. Existing methods leverage Graph Neural Networks with the aid of pre-defined POI graphs to capture such indispensable correlations for modeling user preferences, assuming that the appropriate geographical dependencies among POIs could be pre-determined. However, the pre-defined graph structures are always far from the optimal graph topology due to noise and adaptability issues, which may decrease the expressivity of learned POI representations as well as the credibility of modeling user preferences. In this paper, we propose a novel Adaptive Graph Representation-enhanced Attention Network (AGRAN) for next POI recommendation, which explores the utilization of graph structure learning to replace the pre-defined static graphs for learning more expressive representations of POIs. In particular, we develop an adaptive POI graph matrix and learn it via similarity learning with POI embeddings, automatically capturing the underlying geographical dependencies for representation learning. Afterward, we incorporate the learned representations of POIs and personalized spatial-temporal information with an extension to the self-attention mechanism for capturing dynamic user preferences. Extensive experiments conducted on two real-world datasets validate the superior performance of our proposed method over state-of-the-art baselines.

Functional urban land use recognition integrating multi-source geospatial data and cross-correlations

With the speedy expansion of Internet of Spatial Things, the enormous volume of geospatial big data is produced by the IoT devices. It gives rise to the new challenges for real time geospatial data processing and storing of reliable data in cloud system. The traditional geospatial cloud computing system is not efficient enough to process large volumetric of concurrent geospatial data. Consequently, fog assisted cloud computing environment has come into picture for achieving secure geospatial big data deduplication scheme. In this paper, we introduce a novel scheme GeoBD2 which defines the geo-deduplication structure to build an efficient geospatial bigdata deduplication scheme on fog assisted cloud computing framework. It also regulates which fog node needs to be traversed to investigate duplicate geospatial data rather than to traverse all the fog nodes. This can substantially enhance the efficiency of geospatial big data deduplication in fog assisted cloud environment. It also executes the performance analysis of the proposed scheme. By the experimental results, it is found that the proposed scheme has minimum overhead cost than the existing big data deduplication scheme.

Satellite imagery depicts the Earth’s surface remotely and provides comprehensive information for many applications, such as land use monitoring and urban planning. Existing studies on unsupervised representation learning for satellite images only take into account the images’ geographic information, ignoring human activity factors. To bridge this gap, we propose using the Point-of-Interest (POI) data to capture human factors and designing a contrastive learning-based framework to consolidate the representation of satellite imagery with POI information. Besides, we introduce a season-invariant representation learning model on satellite imagery, considering that human factors are mostly unchanging with respect to seasons. An attention model is designed at last to merge the representations from the geographic, seasonal, and POI perspectives adaptively. On the basis of real-world datasets collected from Beijing, 1 we evaluate our method for predicting socioeconomic indicators. The results show that the representation containing POI information outperforms the geographic representation in estimating commercial activity-related indicators. Our proposed attentional framework can estimate the socioeconomic indicators with R 2 of 0.874 and outperforms the baseline methods. Furthermore, we explore the differences in the representations of satellite images with varying socioeconomic statuses. Finally, we investigate the impact of geographic and POI perspective information in the representation learning process, as well as the effect of satellite imagery on various spatial resolutions.

Multi-modal knowledge graphs representation learning via multi-headed self-attention

Understanding intrinsic patterns and predicting spatiotemporal characteristics of cities require a comprehensive representation of urban neighborhoods. Existing works relied on either inter- or intra-region connectivities to generate neighborhood representations but failed to fully utilize the informative yet heterogeneous data within neighborhoods. In this work, we propose Urban2Vec, an unsupervised multi-modal framework which incorporates both street view imagery and point-of-interest (POI) data to learn neighborhood embeddings. Specifically, we use a convolutional neural network to extract visual features from street view images while preserving geospatial similarity. Furthermore, we model each POI as a bag-of-words containing its category, rating, and review information. Analog to document embedding in natural language processing, we establish the semantic similarity between neighborhood (“document”) and the words from its surrounding POIs in the vector space. By jointly encoding visual, textual, and geospatial information into the neighborhood representation, Urban2Vec can achieve performances better than baseline models and comparable to fully-supervised methods in downstream prediction tasks. Extensive experiments on three U.S. metropolitan areas also demonstrate the model interpretability, generalization capability, and its value in neighborhood similarity analysis.

Urban physical environments are the physical settings and built environments in neighbourhoods and cities which provide places for human activities. Evidence suggests that there are substantial associations between urban physical environments and various health outcomes, e.g. people’s physical activities might be influenced by surrounding physical environments, thereby affecting their health behaviours; more exposure to urban physical environments may benefit human mental health. Street view imagery enables us to capture the landscape at eye-level, making it a promising data source for observing and analysing the realistic dynamics of urban physical environments. Compared with traditional in-person assessments and field observations, street view imagery-based data collection is relatively time-effective and cost-effective. Researchers from epidemiology, psychology, and geography have used street view imagery to quantify the built environment and understand its impacts on public health. To summarize current research trends, this paper systematically reviews the use of street view images for sensing urban environments in public health studies. Specifically, we describe the characteristics of street view imagery and review the methodology for image processing and semantic understanding. We then summarize the challenges that remain for quantifying urban environments in terms of data and methodology. Several future research directions that would benefit public health research and practices are recommended in urban environment research.

This work focuses on using Deep Reinforcement Learning (DRL) to control an autonomous vehicle in the hyper-realistic urban simulation LGSVL. Classical control systems such as MPC maneuver vehicles based on a given trajectory, current velocity, position, distances, and more. Our approach does not pass this information to the DRL agent but only images provided by the camera. Current DRL efforts also exploit similar approaches for autonomous driving, but they are only suitable for small, simple tasks using simple simulations. Our approach consists of two differently trained neural networks (NN), a perceptual NN for representation learning and an actor NN for selecting the correct action. The perception NN will be trained via representation and self-supervised learning to strengthen our DRL agent's understanding of the scene. It can recognize temporal information and the dynamics of a complex environment. This work shows the importance of decoupling the perception and decision (actor) model for autonomous driving. All in all, we could drive autonomously in a hyper-realistic urban simulation using our modular DRL framework. Moreover, our approach also provides a solution for other similar tasks in the field of robotics based on images.

Understanding the role of urban social and physical environment in opioid overdose events using found geospatial data.