In recent decades, traffic problems have become more severe due to population growth in urban areas and the associated number of cars and motorbikes. This has led to an increase in problems related to congestion control, which directly affects citizens: air pollution, fuel consumption, violation of traffic rules, noise pollution, accidents and loss of time. Therefore, intelligent traffic systems are needed to improve the efficiency of traffic flow. However, processing and modelling traffic data is difficult due to the complexity of road networks, spatio-temporal dependencies, and heterogeneous traffic patterns. Here, the quantifiable results of the accuracy of different methods based on deep learning techniques are compared. The best result of each work for traffic flow prediction was selected, which was mainly obtained for a forcecast interval of 15 minutes. A comparison of the results of these models to those of additional traffic flow prediction models can be found in Medina-Salgado et al., 2022.


    Access

    Redirect to the ORKG


    Export, share and cite



    Title :

    Traffic Speed Prediction with Deep Learning Methods - Best results


    Contributors:

    Publication date :

    2022



    Type of media :

    Research Data


    Type of material :

    Electronic Resource


    Language :

    English





    Traffic Speed Prediction with Deep Learning Methods - Best results (v2)

    Ganske, Anette | ORKG Comparisons | 2023

    Free access

    Traffic speed prediction using deep learning method

    Yuhan Jia, / Jianping Wu, / Yiman Du, | IEEE | 2016


    Rainfall‐integrated traffic speed prediction using deep learning method

    Jia, Yuhan / Wu, Jianping / Ben‐Akiva, Moshe et al. | Wiley | 2017

    Free access

    Rainfall-integrated traffic speed prediction using deep learning method

    Jia, Yuhan / Wu, Jianping / Ben-Akiva, Moshe et al. | IET | 2017

    Free access

    Long-Term Urban Traffic Speed Prediction With Deep Learning on Graphs

    Yu, James J. Q. / Markos, Christos / Zhang, Shiyao | IEEE | 2022