A decomposition-based forecasting method with transfer learning for railway short-term passenger flow in holidays

作者：

Highlights：

• We study a decomposition-based forecasting method with transfer learning for railway short-term passenger flow in holidays.

• We propose feature selection and samples filtering methods to generate samples in the Source domain.

• TrAdaBoost is applied to realize the transfer learning.

• The transfer method outperforms the baselines on a real-world data.

摘要

•We study a decomposition-based forecasting method with transfer learning for railway short-term passenger flow in holidays.•We propose feature selection and samples filtering methods to generate samples in the Source domain.•TrAdaBoost is applied to realize the transfer learning.•The transfer method outperforms the baselines on a real-world data.

论文关键词：Railway,Passenger flow forecast,Feature selection,Transfer learning,Sample filtering,Decomposition

论文评审过程：Received 21 March 2021, Revised 11 September 2021, Accepted 14 October 2021, Available online 30 October 2021, Version of Record 9 November 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116102