Modeling train timetables as images: A cost-sensitive deep learning framework for delay propagation pattern recognition

作者：

Highlights：

• A hybrid deep learning model was developed for delay propagation patterns.

• Train timetables were modeled as images.

• A cost-sensitive technique was used to address the data imbalance challenge.

• The proposed model shows satisfactory performance on different situations.

摘要

•A hybrid deep learning model was developed for delay propagation patterns.•Train timetables were modeled as images.•A cost-sensitive technique was used to address the data imbalance challenge.•The proposed model shows satisfactory performance on different situations.

论文关键词：Train delay propagation,Pattern recognition,Train timetables,Images,Imbalanced data

论文评审过程：Received 27 January 2020, Revised 26 December 2020, Accepted 2 April 2021, Available online 6 April 2021, Version of Record 16 April 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.114996