C46: Advances in Spatio Temporal Analytics and GI Technologies
Studying Spatial and Temporal Visitation Patterns of Points of Interest Using SafeGraph Data in Florida
1Florida International University, USA; 2University of Florida, USA
SafeGraph is a commercial provider of massive Point of Interest data including visitation patterns in North America. Although the data source does not share specific travel trajectories, daily and monthly POI visitation numbers for over 160 categories as well as information about from where people visit and what other POI categories they visit are available. This allows analysts to gain insight into travel behavior in a geographic region over time. This study analyzes various aspects of visitation patterns that can be derived from the SafeGraph data set for Florida. Using three Florida major cities, Miami, Orlando, and Jacksonville as study areas, temporal patterns of daily and monthly visit numbers are correlated between various POI categories, and the effect of a short event (Hurricane Irma) on daily visitation numbers around the event is explored. Also, travel distances from home to POIs are compared between different POI categories, and Ordinary Least Squares regression models are used to identify factors associated with increased or decreased travel distance to a specific POI category from home. The study concludes that the aggregated data provided in the SafeGraph platform helps to learn more about travel patterns both in the spatial and temporal domain.
Exploratory Spatiotemporal Language Analysis of Geo-Social Network Data for Identifying Refugee Movements
1University of Salzburg, Österreich; 2Salzburg University of Applied Sciences, Österreich; 3Harvard University, USA
Refugee movements of recent years have caused enormous challenges for relief organisations and public authorities, but especially for refugees themselves. Organisations who have to allocate their resources to regions where large groups of arrivals are expected, struggle to prepare the refugees’ admission, transfer, care, and accommodation in time. Events like the refugee movement 2015/16 in Austria and Germany in the wake of the Syrian civil war have shown that many of these issues are caused by a lack of up-to-date information about logistic requirements. We evaluate different methods to acquire this information by analysing geo-social network data that utilize semantic, spatial, and temporal features. A multimodal analysis of these features leads to information about refugee movement across borders and regions. Approaches based on user trajectories and attempts to identify refugees by their used language showed little promise, whereas using spatiotemporal aggregation and hot spot analysis of keyword-based filtered data allowed us to retrace refugees’ collective movement patterns. Using temporal bins, we were able to detect changes in these patterns caused by external factors such as border closings.
Spatial Operators for Complex Event Processing
1Disy Informationssysteme GmbH, Deutschland; 2FZI - Das Forschungszentrum Informatik, Deutschland
The types of available data have changed in the last decade. While historically data was gathered in batches and distributed as such, e.g. as a database or shapefile, today we are dealing increasingly with real-time data. This data is produced and consumed at the same time and has no pre-known ending. This is most commonly known as streaming data. Traditionally, software for spatial analysis, such as a GIS or spatial data bases, are created and optimized for the batch processing of data. However, the inherent property of streaming data provides unique and new challenges for data-stream processing systems, which are not yet solved. In this paper we propose to enhance systems for handling and analysis of streaming data with spatial operators. We identified Complex Event Processing (CEP) as a promising underlying concept for such a system and use the (open source) self-service IoT toolbox “StreamPipes” as a representative for this. Based upon the GI literature, we identified 6 fundamental types of spatial operators and implemented 32 basic spatial operators in 9 differential groups which can be combined with the existing non-spatial operators for in-depth data analysis.
Extracting Patterns from Large Movement Datasets
1AIT Austrian Institute of Technology, Vienna, Austria; 2University of Salzburg, Austria; 3UNIGIS Lehrende
Extracting useful information from large spatio-temporal datasets is a challenging task that requires suitable visual data representations. Big movement data are particularly hard to visualize since they are prone to visual clutter caused by overlapping and crisscrossing trajectories. Different data aggregation approaches have been developed to address this challenge and provide analysists with better visualizations for data exploration and data-driven hypothesis generation. However, most approaches for extracting patterns, such as mobility graphs or generalized flow maps, cannot handle large input datasets. This paper presents a flow extraction algorithm that can be used in distributed computing environments and thus makes it possible to explore movement patterns in large datasets. We demonstrate its usefulness in a use case exploring maritime vessel movements.