This paper addresses the long-standing challenge of traffic congestion in urban areas exacerbated by population growth and increased vehicular traffic. Accurate traffic flow prediction is crucial for the effective functioning of intelligent transportation systems (ITS) to control traffic and enhance safety. This study focused on exploring various hybrid deep learning models, including LSTM_ARIMA, LSTM_ XGB, and GRU_CNN. Among these, LSTM_ARIMA emerged as the most promising and consistently outperformed the other methods in terms of the prediction accuracy. Experiments conducted on real traffic data demonstrated a significant reduction in the root mean square error and an improvement in the R score when the LSTM_ARIMA model was utilized. The findings of this research contribute to the advancement of traffic prediction methods, with implications for urban planning, traffic optimization, and overall traffic safety enhancement.
Integrating Hybrid Deep Learning for Improving Traffic Flow Prediction
Lect. Notes in Networks, Syst.
International Conference on Smart Systems and Emerging Technologies ; 2024 ; Marrakech, Morocco November 19, 2024 - November 21, 2024
International Conference on Smart Systems and Emerging Technologies ; Chapter : 9 ; 98-107
2025-08-14
10 pages
Article/Chapter (Book)
Electronic Resource
English
Traffic flow prediction method based on hybrid deep learning
European Patent Office | 2023
|Short-term traffic flow prediction using deep learning
SLUB | 2023
|Deep Learning for Short-Term Traffic Flow Prediction
Online Contents | 2016
|