TY - GEN
T1 - A Comparison of Traffic Flow Prediction Methods Based on DBN
AU - Tan, Huachun
AU - Xuan, Xuan
AU - Wu, Yuankai
AU - Zhong, Zhiyu
AU - Ran, Bin
N1 - Publisher Copyright:
© 2016 ASCE.
PY - 2016
Y1 - 2016
N2 - Accurate and real-time traffic flow prediction nowadays shows more and more dependence on big transportation data. Deep learning, a powerful method for feature learning, has turned out to be an effective tool to cope with these explosive data. Recently, deep models, especially unsupervised models like deep belief networks (DBN) and stacked autoencoder (SAE), are being employed into the field of traffic research and have shown great prospect. However, there is still a vacancy in the exploration on comparing the performances of different kinds of deep architectures to find an optimal solution. In this paper, we set up two deep-learning-based traffic flow prediction models for feature extraction and performances comparison: One is a deep belief networks (DBN) based on restricted Boltzmann machines (RBMs) that have Gaussian visible units and binary hidden units, and the other is a DBN based on RBMs with all units being binary. A conclusion is drawn where the former one performs better in traffic flow prediction after a series of experiments.
AB - Accurate and real-time traffic flow prediction nowadays shows more and more dependence on big transportation data. Deep learning, a powerful method for feature learning, has turned out to be an effective tool to cope with these explosive data. Recently, deep models, especially unsupervised models like deep belief networks (DBN) and stacked autoencoder (SAE), are being employed into the field of traffic research and have shown great prospect. However, there is still a vacancy in the exploration on comparing the performances of different kinds of deep architectures to find an optimal solution. In this paper, we set up two deep-learning-based traffic flow prediction models for feature extraction and performances comparison: One is a deep belief networks (DBN) based on restricted Boltzmann machines (RBMs) that have Gaussian visible units and binary hidden units, and the other is a DBN based on RBMs with all units being binary. A conclusion is drawn where the former one performs better in traffic flow prediction after a series of experiments.
KW - Deep Belief Networks (DBN)
KW - Deep learning
KW - Restricted Boltzmann machines (RBM)
KW - Traffic flow prediction
UR - http://www.scopus.com/inward/record.url?scp=84979790091&partnerID=8YFLogxK
U2 - 10.1061/9780784479896.026
DO - 10.1061/9780784479896.026
M3 - Conference contribution
AN - SCOPUS:84979790091
T3 - CICTP 2016 - Green and Multimodal Transportation and Logistics - Proceedings of the 16th COTA International Conference of Transportation Professionals
SP - 273
EP - 283
BT - CICTP 2016 - Green and Multimodal Transportation and Logistics - Proceedings of the 16th COTA International Conference of Transportation Professionals
A2 - Ge, Ying-En
A2 - Wang, Xiaokun
A2 - Zhang, Yu
A2 - Huang, Youfang
PB - American Society of Civil Engineers (ASCE)
T2 - 16th COTA International Conference of Transportation Professionals: Green and Multimodal Transportation and Logistics, CICTP 2016
Y2 - 6 July 2016 through 9 July 2016
ER -