arrow
Return

A multiscale-based model for train delay prediction considering spatio-temporal correlations

delete2025-10-27
delete0
delete
OA
AI
J
Jianqing Wu
X
Xukai Xiao
B
Bo Du *
J
Jun Shen
Z
Zhan Wu
Y
Yihui Wang
X
Xiangchun Yu
S
Shuxin Yang
DOI:10.1016/j.trc.2025.105417delete
deleteOriginal
deleteShare
deleteSave
View PDF
Abstract

Abstract

En 中文
• Adaptive multi-scale segmentation module was designed to capture temporal dynamics. • The GA-MTSC framework integrated linear attention and adaptive graph convolution. • MTSC reweights multi-scale outputs to capture nonlinear and periodic patterns. • Experiment with real-world data showed superior performance of the proposed method.
Keywords:
Train delay prediction
Frequency domain
Spatio-temporal correlation
Multi time scales
AI Summary

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.

Journal

T
transportation research part c: emerging technologies
IF:
0
Papers:
196
Citations:
0

Organization

B
Beijing Jiaotong University
Scholars:
2.2W
Papers: 1.7W
Citations: 1.2W
U
University of Wollongong
Scholars:
1.3W
Papers: 1.6W
Citations: 2.8W
J
Jiangxi University of Science and Technology
Scholars:
3.9K
Papers: 1.2K
Citations: 7.2K
G
Griffith University
Scholars:
1.5W
Papers: 1.6W
Citations: 2.5W
researcher View more organizations