Traffic speed prediction has implications for urban planning, congestion reduction, and intelligent control systems. To maintain a uniform traffic speed and to avoid issues related to traffic, an accurate traffic speed forecast can help in supplying significant information. The capacity to forecast short-term traffic speed is a fundamental part of both Intelligent Transportation System (ITS) and the Internet of Vehicles (IoV). To achieve better accuracy in predicting short-term traffic speed, we introduced a GSA-Bi-LSTM model by optimizing the Bi-directional Long Short-Term Memory (Bi-LSTM) network prediction framework with Gravitational Search Algorithm (GSA) due to its features of fast convergence, great reliability and significant global search ability of parameters. The utilization of the GSA optimization technique is employed to optimize the hyperparameters of the Bi-LSTM model. By making use of the bidirectional properties of Bi-LSTM layers, the model’s architecture aims to enhance prediction accuracy and effectively capture the intricate patterns present in the input data. From the analysis of the experimental results, it becomes evident that the convenience provided by our proposed GSA-Bi-LSTM model surpasses that of conventional models in terms of evaluation metrics. Additionally, it is also noted that GSA has superior optimization capabilities than Particle Swarm Optimization (PSO) in terms of optimizing the Bi-LSTM approach for traffic speed forecasting.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Bi-directional Long Short Term Memory Neural Network for Short-Term Traffic Speed Prediction Using Gravitational Search Algorithm


    Additional title:

    Int. J. ITS Res.


    Contributors:


    Publication date :

    2024-08-01


    Size :

    12 pages




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English