Based on deep learning methods, especially long short-term memory (LSTM) neural networks, short-term traffic forecasting has achieved explosive growth. This study proposes the Bi-LSTM model to effectively predict travel time. In order to validate the effectiveness of the proposed stacked LSTM, we used 9-day toll station entry and exit data from the expressways of Guangdong province with an updating frequency of 5 min. The experimental result indicates that excessive depths of the model will lead to the increase of loss values. Moreover, the stability of data will affect the prediction accuracy. In addition, compared with other machine learning methods, as well as different topologies of neural networks, the stacked Bi-LSTM neural network has advantages of reliability, accuracy, and stability, which could facilitate travel time prediction.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Long Short-Term Memory Neural Network for Travel Time Prediction of Expressways Using Toll Station Data


    Contributors:
    Chen, Deqi (author) / Yan, Xuedong (author) / Li, Shurong (author) / Wang, Liwei (author) / Liu, Xiaobing (author)

    Conference:

    20th COTA International Conference of Transportation Professionals ; 2020 ; Xi’an, China (Conference Cancelled)


    Published in:

    CICTP 2020 ; 73-85


    Publication date :

    2020-08-12




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English





    Long Short-Term Memory Neural Network for Traffic Speed Prediction of Urban Expressways Using Floating Car Data

    Chen, Deqi / Yan, Xuedong / Li, Shurong et al. | British Library Conference Proceedings | 2022