Citation

Lim, Le Xiang and Tee, Connie and Goh, Michael Kah Ong (2024) Enhancing Traffic Analysis and Prediction through A Hybrid LSTM-ARIMA Model. In: 2024 International Conference on Information Management and Technology (ICIMTech), 28-29 August 2024, Bali, Indonesia.

Text Enhancing Traffic Analysis and Prediction through A Hybrid LSTM-ARIMA Model.pdf - Published Version Restricted to Repository staff only Download (7MB)

Abstract

Innovative methods for analyzing traffic are crucial for modernizing transportation infrastructure, particularly in highway planning. This study explores the integration of computer vision techniques, emphasizing image processing and machine learning, for real-time highway traffic analysis. Given the limitations of traditional methods like slow speed and inaccuracy, the importance of computer visionbased real-time analysis can never be exaggerated. In this paper, a hybrid LSTM-ARIMA model is proposed for accurate highway traffic flow estimation by exploiting modern deep learning techniques. This model fuses the advantages of Long Short-Term Memory (LSTM) with those of Autoregressive Integrated Moving Average (ARIMA) to improve forecasting accuracy. Based on the experimental findings, LSTM-ARIMA is more effective than both standalone LSTM and ARIMA approaches since it has a stacking MAE of 0.2662 in case of outgoing traffic as compared to 0.5385 for incoming traffic. These findings demonstrate strong evidence that our proposed approach could revolutionize traffic analysis and management to aid transportation authorities and infrastructure developers in making informed choices when planning their cities or municipalities.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Traffic analysis, computer vision
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	07 Feb 2025 02:45
Last Modified:	07 Feb 2025 02:46
URII:	http://shdl.mmu.edu.my/id/eprint/13394

Downloads

Edit (login required)