Road accidents are one of the major causes of death and injury all over the world, hence an enormous demand to conduct research using sophisticated algorithms and techniques for evaluating and predicting traffic accidents and identifying the key factors causing road accidents. Rapid urbanization has led to an increase in the number of vehicles on the road, resulting in tragic traffic accidents that have caused numerous casualties and significant financial losses. Finding a solution to the challenge of allocating resources to make transportation safer is the ultimate goal of road accident prediction research, which aims to save lives. To improve the accuracy of traffic accident risk forecasting, we use four deep learning prediction models: 1) Deep Neural Network (DNN); 2) Convolutional Neural Network (CNN); 3) Gated Recurrent Unit (GRU); 4) Long Short-Term Memory (LSTM). Lastly, we apply different performance indicators such as MSE, Bias, MAE, RMSE and MAPE through a comparative study to achieve the best predictive road accident model. When compared to other deep learning methods, the DNN model outperforms them all when it comes to predicting the seriousness of incidents involving buses, cars, and trucks. The GRU is the best neural network model available, second only to DNN. An intelligent traffic management system can incorporate the suggested method to improve organization control and traffic predictions.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Predicting the Risk of Citywide Traffic Road Accidents Using Deep Learning Techniques


    Weitere Titelangaben:

    Lect. Notes in Networks, Syst.


    Beteiligte:
    Kaiser, M. Shamim (Herausgeber:in) / Xie, Juanying (Herausgeber:in) / Rathore, Vijay Singh (Herausgeber:in) / Poonkuzhali, S. (Autor:in) / Rajesh Kannan, G. (Autor:in)

    Kongress:

    International Conference on Information and Communication Technology for Competitive Strategies ; 2024 ; Jaipur, India December 19, 2024 - December 21, 2024



    Erscheinungsdatum :

    29.07.2025


    Format / Umfang :

    10 pages





    Medientyp :

    Aufsatz/Kapitel (Buch)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch





    A Deep Learning Approach to the Citywide Traffic Accident Risk Prediction

    Ren, Honglei / Song, You / Wang, Jingwen et al. | IEEE | 2018


    A Deep Learning Approach to the Citywide Traffic Accident Risk Prediction

    Ren, Honglei / Song, You / Wang, Jingwen et al. | ArXiv | 2017

    Freier Zugriff


    A survey on citywide traffic estimation techniques

    Abdulla, Dudekula / Ramu, Gandikota / Mamatha, N | IEEE | 2017