ABSTRACT
Though GPS-based human trajectory data have been commonly used in travel surveys and human mobility studies, missing data or data gaps that are intrinsically relevant to research reliability remain a critical and challenging issue. This study proposes a novel framework for imputing data gaps based on frequent-pattern mining and time geography, which allows for considering spatio-temporal travel restrictions during imputation by evaluating the spatio-temporal topology relations between the space-time prisms of gaps and corresponding frequent activities or trips. For the validation, the proposed framework is applied to raw GPS trajectories that were collected from 139 participants in Switzerland. In the case study, the temporal and spatio-temporal gaps are artificially generated by randomly choosing activities and trips from the trajectory data. Through comparing the mobility indicators (i.e. duration and distance) calculated from raw data, imputed data, and data with gaps, we quantitatively evaluate the performance of the proposed method in terms of Pearson correlation coefficients and deviation. We further compare the framework with the shortest path interpolation method based on the generated spatio-temporal gaps. The comparison results demonstrate the performance and advantage of the proposed method in imputing gaps from GPS-based human movement data.
Acknowledgments
The authors sincerely appreciate all valuable and insightful comments and suggestions from the anonymous reviewers and editor, which remarkably improve the quality of this article.
Disclosure statement
No potential conflict of interest was reported by the authors.
Data and codes availability statement
The codes and mocked data sample from GeoLife trajectory dataset that support the findings of this study are available in ‘figshare.com’ at the private link: https://doi.org/10.6084/m9.figshare.12016341.
All participants’ trajectory data cannot be made publicly available due to confidentiality agreements with Swiss Federal Railways SBB.
Notes
Additional information
Funding
Notes on contributors
Pengxiang Zhao
Pengxiang Zhao received his Ph.D. degree in Cartography and Geographic Information Engineering from Wuhan University in 2015. He joined the Institute of Cartography and Geoinformation, the Swiss Federal Institute of Technology (ETH) Zurich as a postdoctoral fellow in 2018. His research interests include geographic information science, urban mobility, and geospatial data analysis and mining.
David Jonietz
David Jonietz is Principal Machine Learning & Artificial Intelligence Engineer at HERE Technologies in Zurich. He is also a Scientific Advisor at the Institute of Advanced Research in Artificial Intelligence (IARAI) in Vienna. Previously, David was working as PostDoc Researcher at the Swiss Federal Institute of Technology (ETH) Zurich and the University of Heidelberg. In 2016, he received his Dr. rer. nat. in Geoinformatics from the University of Augsburg. His research interests lie mainly in the areas of machine learning, geoinformatics and urban transport systems research.
Martin Raubal
Martin Raubal is professor of geoinformation engineering at the Swiss Federal Institute of Technology (ETH) Zurich. He was previously associate professor and Vice-Chair at the Department of Geography, University of California, Santa Barbara. Martin received his Dr. techn. in geoinformation from Vienna University of Technology in 2001 with honors. His research interests lie in the areas of mobility & energy, more specifically in mobile GIS & LBS, spatial cognitive engineering, mobile eye-tracking, and GIS for renewable energy analysis. Martin serves on the editorial boards of Transactions in GIS, Journal of Spatial Information Science, Annals of the American Association of Geographers, and Spatial Cognition and Computation. He has authored and coauthored more than 150 books and research papers published in refereed journals and conference proceedings.