TY - JOUR
T1 - Deep and Embedded Learning Approach for Traffic Flow Prediction in Urban Informatics
AU - Zheng, Zibin
AU - Yang, Yatao
AU - Liu, Jiahao
AU - Dai, Hong Ning
AU - Zhang, Yan
N1 - Funding Information:
Manuscript received March 25, 2018; revised August 6, 2018, December 5, 2018 and February 18, 2019; accepted March 11, 2019. Date of publication April 22, 2019; date of current version October 2, 2019. This work was supported by the National Key Research and Development Program under Grant 2016YFB1000101, the National Natural Science Foundation of China under Grant 61722214, Macao Science and Technology Development Fund under Grant 0026/2018/A1, the Guangdong Province Universities and Colleges Pearl River Scholar Funded Scheme (2016), the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Skłodowska-Curie Agreement 824019, the Sichuan Science and Technology Program under Grant 2019YFH0033, and the Program for Guangdong Introducing Innovative and Entrepreneurial Teams under Grant 2016ZT06D211. The Associate Editor for this paper was S.-H. Kong. (Corresponding author: Yan Zhang.) Z. Zheng, Y. Yang, and J. Liu are with the School of Data and Computer Science, Sun Yat-sen University, Guangzhou 510006, China, and also with the National Engineering Research Center of Digital Life, Sun Yat-sen University, Guangzhou 510006, China (e-mail: [email protected]; [email protected]; [email protected]).
Publisher Copyright:
© 2000-2011 IEEE.
PY - 2019/10
Y1 - 2019/10
N2 - Traffic flow prediction has received extensive attention recently, since it is a key step to prevent and mitigate traffic congestion in urban areas. However, most previous studies on traffic flow prediction fail to capture fine-grained traffic information (like link-level traffic) and ignore the impacts from other factors, such as route structure and weather conditions. In this paper, we propose a deep and embedding learning approach (DELA) that can help to explicitly learn from fine-grained traffic information, route structure, and weather conditions. In particular, our DELA consists of an embedding component, a convolutional neural network (CNN) component and a long short-term memory (LSTM) component. The embedding component can capture the categorical feature information and identify correlated features. Meanwhile, the CNN component can learn the 2-D traffic flow data while the LSTM component has the benefits of maintaining a long-term memory of historical data. The integration of the three models together can improve the prediction accuracy of traffic flow. We conduct extensive experiments on realistic traffic flow dataset to evaluate the performance of our DELA and make comparison with other existing models. The experimental results show that the proposed DELA outperforms the existing methods in terms of prediction accuracy.
AB - Traffic flow prediction has received extensive attention recently, since it is a key step to prevent and mitigate traffic congestion in urban areas. However, most previous studies on traffic flow prediction fail to capture fine-grained traffic information (like link-level traffic) and ignore the impacts from other factors, such as route structure and weather conditions. In this paper, we propose a deep and embedding learning approach (DELA) that can help to explicitly learn from fine-grained traffic information, route structure, and weather conditions. In particular, our DELA consists of an embedding component, a convolutional neural network (CNN) component and a long short-term memory (LSTM) component. The embedding component can capture the categorical feature information and identify correlated features. Meanwhile, the CNN component can learn the 2-D traffic flow data while the LSTM component has the benefits of maintaining a long-term memory of historical data. The integration of the three models together can improve the prediction accuracy of traffic flow. We conduct extensive experiments on realistic traffic flow dataset to evaluate the performance of our DELA and make comparison with other existing models. The experimental results show that the proposed DELA outperforms the existing methods in terms of prediction accuracy.
KW - deep learning
KW - embedding neural networks
KW - traffic flow prediction
KW - Urban informatics
UR - http://www.scopus.com/inward/record.url?scp=85075535431&partnerID=8YFLogxK
U2 - 10.1109/TITS.2019.2909904
DO - 10.1109/TITS.2019.2909904
M3 - Journal article
SN - 1524-9050
VL - 20
SP - 3927
EP - 3939
JO - IEEE Transactions on Intelligent Transportation Systems
JF - IEEE Transactions on Intelligent Transportation Systems
IS - 10
ER -