Browsing by Subject "trajectory"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Arrested (during) Development: County Young Adult Arrest Rate Patterning across Space and Time
    (2023) Henderson, Brit
    This three-paper dissertation project includes an examination of county-level young adult arrest rate trajectories over several decades and their association with various measures of county ecological context. Then, I assess the extent to which there is spatial patterning in young adult arrest rates, above and beyond ecological measures of place. Finally, I offer an initial exploration into the connection between county young adult arrest rates and county mental health outcomes.
  • Thumbnail Image
    Item
    DynamoVis - Dynamic Visualization of Animal Movement Data
    (2018-06-01) Somayeh Dodge; Glenn Xavier; Wing Yi Wong; sdodge@umn.edu; Dodge, Somayeh; University of Minnesota, Department of Geography, Environment, and Society
    Exploring movement, as an important aspect of spatiotemporal processes, has gained new momentum from the availability of large spatiotemporal datasets. This has given rise to the development of new exploratory and analytical techniques to generate new insight into dynamic processes and the spatiotemporal context in which they operate. This study develops a new dynamic visualization tool, called ``DYNAMOVis: Dynamic Visualization of Movement'', developed for the exploratory analysis of movement in relation to the environment and geographic context. DYNAMOVis applies visual variables such as point and line width, color, and directional vector to visualize movement tracks in their attribute space (e.g. movement parameters and context attributes). Using real case studies from Movement Ecology, we show how hybrid and dynamic visualizations can strengthen spatiotemporal research by facilitating data exploration, generating new hypotheses, discovery of patterns and dependencies, as well as promoting interdisciplinary research collaborations.
  • Thumbnail Image
    Item
    PIMMLI: Predictive In-Memory Multi-Level Indexing for Distributed Trajectory Streams
    (2021-05) Samad, Abdul
    The popularity of location-based social media and GPS-enabled mobile devices has produced a large amount of streaming trajectory data. Each streaming trajectory consists of the sequence of positions that a moving object occupies in time and is generated in an online fashion, coming at high speed. Disciplines such as social networking, urban planning, ecology, and epidemiology have great interest in querying this type of data. However, the large volume of streaming trajectories poses scalability challenges that can be addressed by efficient indexing structures and in-memory distributed architectures such as Spark. Despite this, no streaming trajectory query processing algorithm has been proposed that uses indices and distributed architectures to tackle this large-scale problem. To address this, we propose a novel in-memory predictive multi-level indexing technique, called PIMMLI, that leverages the distributed Spark Streaming framework to process spatio-temporal queries on streaming trajectories in an efficient manner. We evaluated the effectiveness of PIMMLI on 3 real-life large-scale datasets. These experiments showed that PIMMLI had an average improvement of 3.5X in total query execution and indexing time over DITA, an existing state-of-the-art batch processing algorithm for spatio-temporal querying on trajectories, and of 34.09X in query execution time over an approach that uses no indices.
  • Thumbnail Image
    Item
    Scalable GPU Algorithms for Similarity Query Processing and Clustering on Trajectory Data
    (2019-11) Mustafa, Hamza
    With the increasing prevalence of location sensor devices like GPS, it has been possible to collect large datasets of a special type of spatio-temporal data called trajectory data. A trajectory is a discrete sequence of positions that a moving object occupies in space as time passes. Such large datasets enable researchers to study the behavior of the objects describing these movements by issuing spatial queries such as trajectory similarity queries and trajectory clustering. Top-K trajectory similarity queries retrieve the K most similar trajectories to a given query trajectory. This query has applications in many areas, such as urban planning, ecology and social networking; however, this query is computationally expensive. In this work, we introduce a new parallel top-K trajectory similarity query technique for GPUs, FastTopK, to deal with these challenges. Our experiments on two large real-life datasets showed that FastTopK produces on average 107.96X smaller candidate result sets, and 3.36X faster query execution times than the existing state of-the-art technique, TKSimGPU. The second type of trajectory query covered in this work is trajectory clustering. One important algorithm for clustering is DBSCAN, which is especially useful for finding clusters of arbitrary shapes. As opposed to other clustering techniques, like K-means, it does not require the number of clusters to be specified as an input parameter, and it is highly robust to outliers. However, DBSCAN has a worst-case quadratic time complexity that makes it difficult to handle large dataset sizes. To address this problem, several works have been proposed that exploit the massive parallelism of GPUs for DBSCAN clustering of point data. Nonetheless, none of these works have been experimentally compared against each other and none have been extended to cluster trajectory data. In this thesis, we review the existing GPU algorithms for DBSCAN clustering on point data and conduct the first experimental study comparing these GPU algorithms using three real-world datasets to identify the best performing algorithm. Our results show that G-DBSCAN is the fastest being up to 969X faster than CPU DBSCAN on 128K points, while CUDA-DClust is the best performing GPU algorithm in terms of execution time and memory requirements, performing 53X faster than CPU DBSCAN while taking up to 166X less memory than G-DBSCAN. Lastly, we use the work of our analysis of all the existing GPU-based DBSCAN clustering algorithms on point data to develop a new GPU-based trajectory clustering algorithm, GTRACLUS. Our experiments show that on two real world datasets, trajectory clustering can be made more than 20X faster than the CPU-based TRACLUS by using the proposed GTRACLUS algorithm.