STANN : A Spatio-Temporal Attentive Neural Network for Traffic Prediction
Research output: Journal Publications and Reviews (RGC: 21, 22, 62) › 21_Publication in refereed journal › peer-review
Author(s)
Related Research Unit(s)
Detail(s)
Original language | English |
---|---|
Pages (from-to) | 4795-4806 |
Journal / Publication | IEEE Access |
Volume | 7 |
Online published | 18 Dec 2018 |
Publication status | Published - 2019 |
Link(s)
Abstract
Recently, traffic prediction based on deep learning methods has attracted much attention. However, there still exist two major challenges, namely, dynamic spatio-temporal dependencies among network-wide links and long-term traffic prediction for next few hours. To address these two challenges, this paper proposes a Spatio-Temporal Attentive Neural Network (STANN) for the network-wide and long-term traffic prediction. STANN captures the spatial-temporal dependencies based on the encoder-decoder architecture with the attention mechanisms. In the encoder, STANN learns the spatio-temporal dependencies from historical traffic series using a recurrent neural network (RNN) with long short-term memory (LSTM) units, in which a new spatial attention model is developed to consider the contribution of each link to the network-wide prediction. In the decoder, STANN exploits another RNN with LSTM units and a temporal attention model to select relevant and important historical spatio-temporal dependencies from the encoder for long-term traffic prediction. Finally, we conduct extensive experiments to evaluate STANN on three real-world traffic datasets. Experimental results show that STANN is significantly better than other state-of-the-art models.
Research Area(s)
- Spatio-temporal data, deep neural network, attention mechanism, traffic prediction
Citation Format(s)
STANN : A Spatio-Temporal Attentive Neural Network for Traffic Prediction. / HE, Zhixiang; CHOW, Chi-Yin; ZHANG, Jia-Dong.
In: IEEE Access, Vol. 7, 2019, p. 4795-4806.Research output: Journal Publications and Reviews (RGC: 21, 22, 62) › 21_Publication in refereed journal › peer-review