TY - JOUR
T1 - Long-Short Term Spatiotemporal Tensor Prediction for Passenger Flow Profile
AU - Li, Ziyue
AU - Yan, Hao
AU - Zhang, Chen
AU - Tsung, Fugee
N1 - Funding Information:
Manuscript received March 16, 2020; accepted June 4, 2020. Date of publication June 25, 2020; date of current version July 3, 2020. This letter was recommended for publication by Associate Editor Ettore Lanzarone and Editor Jingang Yi upon evaluation of the reviewers’ comments. This work was supported in part by Hong Kong RGC GRF under Grants 16203917 and 16201718, in part by NSFC under Grants 71931006, 71901131, and 71932006, in part by NSF CMMI under Grant 1922739, and in part by NSF under Grant DMS-1830363 and ITT/007/19GP. (H. Yan and C. Zhang contributed equally to this work.) (Corresponding author: Ziyue Li.) Ziyue Li is with the Department of Industrial Engineering and Decision Analytics, Hong Kong University of Science and Technology, Hong Kong (e-mail: zlibn@connect.ust.hk).
Funding Information:
The authors would like to show their great appreciation to the Metro Corporation for sharing this passenger flow data, and in the protection of privacy, all data have been desensitized. They also give special thanks to Dr. Qibin Zhao for sharing the scripts of their methods.
Publisher Copyright:
© 2016 IEEE.
PY - 2020/10
Y1 - 2020/10
N2 - Spatiotemporal data are very common in many applications, such as manufacturing systems and transportation systems. Given the intrinsic complex spatial and temporal correlations of such data, short-term and long-term prediction for spatiotemporal data is often very challenging. Most of the traditional statistical models fail to preserve innate features in data alongside their complex correlations. In this paper, we focus on a tensor-based prediction method and propose several practical techniques to improve both long-term and short-term prediction accuracy. For long-term prediction, we propose the 'tensor decomposition + 2-Dimensional Auto-Regressive Moving Average (2D-ARMA)' model, and an effective way to update prediction in real-time; For short-term prediction, we propose to conduct tensor completion based on tensor clustering to avoid oversimplification and ensure accuracy. A case study based on the metro passenger flow data is conducted to demonstrate the improved performance.
AB - Spatiotemporal data are very common in many applications, such as manufacturing systems and transportation systems. Given the intrinsic complex spatial and temporal correlations of such data, short-term and long-term prediction for spatiotemporal data is often very challenging. Most of the traditional statistical models fail to preserve innate features in data alongside their complex correlations. In this paper, we focus on a tensor-based prediction method and propose several practical techniques to improve both long-term and short-term prediction accuracy. For long-term prediction, we propose the 'tensor decomposition + 2-Dimensional Auto-Regressive Moving Average (2D-ARMA)' model, and an effective way to update prediction in real-time; For short-term prediction, we propose to conduct tensor completion based on tensor clustering to avoid oversimplification and ensure accuracy. A case study based on the metro passenger flow data is conducted to demonstrate the improved performance.
KW - Intelligent transportation system
KW - big data in robotics and automation
KW - probability and statistical methods
UR - http://www.scopus.com/inward/record.url?scp=85088088594&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85088088594&partnerID=8YFLogxK
U2 - 10.1109/LRA.2020.3004785
DO - 10.1109/LRA.2020.3004785
M3 - Article
AN - SCOPUS:85088088594
SN - 2377-3766
VL - 5
SP - 5010
EP - 5017
JO - IEEE Robotics and Automation Letters
JF - IEEE Robotics and Automation Letters
IS - 4
M1 - 9126133
ER -