TY - JOUR
T1 - MS-Net
T2 - Multi-Source Spatio-Temporal Network for Traffic Flow Prediction
AU - Fang, Shen
AU - Prinet, Veronique
AU - Chang, Jianlong
AU - Werman, Michael
AU - Zhang, Chunxia
AU - Xiang, Shiming
AU - Pan, Chunhong
N1 - Publisher Copyright:
© 2000-2011 IEEE.
PY - 2022/7/1
Y1 - 2022/7/1
N2 - Predicting urban traffic flow is a challenging task, due to the complicated spatio-temporal dependencies on traffic networks. Urban traffic flow usually has both short-term neighboring and long-term periodic temporal dependencies. It is also noticed that the spatial correlations over different traffic nodes are both local and non-local. What's more, the traffic flow is affected by various external factors. To capture the non-local spatial correlations, we propose a Dilated Attentional Graph Convolution (DAGC). The DAGC utilizes a dilated graph convolution kernel to expand the nodes' receptive field and exploit multi-order neighborhood. Technically, the lower-order neighborhood corresponds to local spatial dependencies, while the higher-order neighborhood corresponds to non-local spatial dependencies between nodes. Based on DAGC, a Multi-Source Spatio-Temporal Network (MS-Net) is designed, which suffices to integrate long-range historical traffic data as well as multi-modal external information. MS-Net consists of four components: a spatial feature extraction module, a temporal feature fusion module, an external factors embedding module, and a multi-source data fusion module. Extensive experiments on three real traffic datasets demonstrates that the proposed model performs well on both the public transportation networks, road networks, and can handle large-scale traffic networks in particular the Beijing bus network which has more than 4,000 traffic nodes.
AB - Predicting urban traffic flow is a challenging task, due to the complicated spatio-temporal dependencies on traffic networks. Urban traffic flow usually has both short-term neighboring and long-term periodic temporal dependencies. It is also noticed that the spatial correlations over different traffic nodes are both local and non-local. What's more, the traffic flow is affected by various external factors. To capture the non-local spatial correlations, we propose a Dilated Attentional Graph Convolution (DAGC). The DAGC utilizes a dilated graph convolution kernel to expand the nodes' receptive field and exploit multi-order neighborhood. Technically, the lower-order neighborhood corresponds to local spatial dependencies, while the higher-order neighborhood corresponds to non-local spatial dependencies between nodes. Based on DAGC, a Multi-Source Spatio-Temporal Network (MS-Net) is designed, which suffices to integrate long-range historical traffic data as well as multi-modal external information. MS-Net consists of four components: a spatial feature extraction module, a temporal feature fusion module, an external factors embedding module, and a multi-source data fusion module. Extensive experiments on three real traffic datasets demonstrates that the proposed model performs well on both the public transportation networks, road networks, and can handle large-scale traffic networks in particular the Beijing bus network which has more than 4,000 traffic nodes.
KW - Graph convolution
KW - artificial intelligence
KW - deep attention mechanism
KW - deep learning
KW - traffic flow prediction
KW - traffic network
UR - http://www.scopus.com/inward/record.url?scp=85103278771&partnerID=8YFLogxK
U2 - 10.1109/tits.2021.3067024
DO - 10.1109/tits.2021.3067024
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
AN - SCOPUS:85103278771
SN - 1524-9050
VL - 23
SP - 7142
EP - 7155
JO - IEEE Transactions on Intelligent Transportation Systems
JF - IEEE Transactions on Intelligent Transportation Systems
IS - 7
ER -