Article: Spatial-temporal attention-based seq2seq framework for short-term travel time prediction Journal: International Journal of Bio-Inspired Computation (IJBIC) 2022 Vol.20 No.1 pp.23 - 37 Abstract: Short-term travel time prediction is an important research direction of the intelligent transportation system. However, due to the complex temporal and spatial correlation of the traffic data, the more accurate and efficient the prediction of the short-term travel time, the harder it is to achieve. In order to conquer the problems and accelerate the construction of smart cities, this study first proposes the seq2seq model based on LSTM network with strong time series processing capability and can avoid manual feature extraction. Then, a seq2seq model based on multi-level attention mechanism is proposed. Finally, the method proposed is evaluated and compared with the traditional prediction methods through experimental simulations based on the real traffic dataset. The results verify the accuracy and effectiveness of the proposed models, demonstrate that the utilisation of multi-level attention mechanism can effectively integrate temporal and spatial features of the dataset, and improve the efficiency of the original seq2seq model. Inderscience Publishers - linking academia, business and industry through research

Title: Spatial-temporal attention-based seq2seq framework for short-term travel time prediction

Authors: Ningqing Zhang; Fei Wang; Xiong Chen; Tong Zhao; Qi Kang

Addresses: Fudan University, No. 220, Handan Road, Yangpu District, Shanghai, China ' Shanghai Spaceflight Precision Machinery Institute, No. 1, Guide Road, Songjiang District, Shanghai, Shanghai 201600, China ' Fudan University, No. 220, Handan Road, Yangpu District, Shanghai, China ' Fudan University, No. 220, Handan Road, Yangpu District, Shanghai, China ' Tongji University, No. 1239, Siping Road, Yangpu District, Shanghai, China

Abstract: Short-term travel time prediction is an important research direction of the intelligent transportation system. However, due to the complex temporal and spatial correlation of the traffic data, the more accurate and efficient the prediction of the short-term travel time, the harder it is to achieve. In order to conquer the problems and accelerate the construction of smart cities, this study first proposes the seq2seq model based on LSTM network with strong time series processing capability and can avoid manual feature extraction. Then, a seq2seq model based on multi-level attention mechanism is proposed. Finally, the method proposed is evaluated and compared with the traditional prediction methods through experimental simulations based on the real traffic dataset. The results verify the accuracy and effectiveness of the proposed models, demonstrate that the utilisation of multi-level attention mechanism can effectively integrate temporal and spatial features of the dataset, and improve the efficiency of the original seq2seq model.

Keywords: attention mechanism; LSTM; seq2seq framework; short-term travel time prediction; intelligent transportation system; ITS.

DOI: 10.1504/IJBIC.2022.126277

International Journal of Bio-Inspired Computation, 2022 Vol.20 No.1, pp.23 - 37

Received: 21 Feb 2022
Accepted: 07 May 2022
Published online: 18 Oct 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Spatial-temporal attention-based seq2seq framework for short-term travel time prediction

Keep up-to-date