Wang, Y. and Mcarthur, D. P. (2017) Detecting Stops from GPS Trajectories: A Comparison of Different GPS Indicators for Raster Sampling Methods. GeoComputation 2017, Leeds, UK, 04-07 Sep 2017.
|
Text
146519.pdf - Accepted Version 1MB |
Publisher's URL: http://www.geocomputation.org/2017/
Abstract
With the increasing prevalence of GPS tracking capabilities on smartphones, GPS trajectories have proven to be useful for an extensive range of research topics. Stop detection, which estimates activity locations, is fundamental for organizing GPS trajectories into semantically meaningful journeys. With previous methods overwhelmingly dependent on thresholds, contextual information or a pre-understanding of the GPS records, this paper addresses the challenge by contributing a ‘top-down’ raster sampling method which samples pre-calculated GPS indicators and clusters the raster cells with significantly different values as stops. We report a comparison of a set of precalculated GPS indicators with two baseline methods. By referencing a ground truth travel dairy, the raster sampling method demonstrates good and reliable capabilities on producing high accuracy, low redundancy and close proximity to the ground truth in three distinct travel use cases. This further indicates a good generic stop detection method.
Item Type: | Conference or Workshop Item |
---|---|
Status: | Published |
Refereed: | No |
Glasgow Author(s) Enlighten ID: | Wang, Dr Yang and Mcarthur, Dr David |
Authors: | Wang, Y., and Mcarthur, D. P. |
College/School: | College of Social Sciences > School of Social and Political Sciences > Urban Studies |
Copyright Holders: | Copyright © 2017 The Authors |
First Published: | First published in GeoComputation 2017 |
Publisher Policy: | Reproduced with the permission of the Editor |
Related URLs: |
University Staff: Request a correction | Enlighten Editors: Update this record