An integrated feature learning approach using deep learning for travel time prediction

作者：

Highlights：

• Integrated supervised & unsupervised algorithm is proposed for travel time prediction.

• Feature enriching algorithms have been applied to increase feature learnability.

• A SAE with dropout for extracting information and increasing robustness is developed.

• A deep FF MLP for travel time prediction using encoded features have been developed.

摘要

•Integrated supervised & unsupervised algorithm is proposed for travel time prediction.•Feature enriching algorithms have been applied to increase feature learnability.•A SAE with dropout for extracting information and increasing robustness is developed.•A deep FF MLP for travel time prediction using encoded features have been developed.

论文关键词：Deep learning,Stacked autoencoder,Transportation,Travel time prediction,Deep neural network,Spatiotemporal analysis

论文评审过程：Received 3 April 2019, Revised 29 July 2019, Accepted 5 August 2019, Available online 6 August 2019, Version of Record 8 August 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.112864