An efficient realization of deep learning for traffic data imputation

•The model denoising stacked autoencoders for traffic data imputation is proposed.•The model performance changes with temporal and spatial factors.•An efficient model realization with hierarchically training algorithm is developed. Traffic data provide the basis for both research and applications in...

Full description

Saved in:
Bibliographic Details
Published in:Transportation research. Part C, Emerging technologies Emerging technologies, 2016-11, Vol.72, p.168-181
Main Authors: Duan, Yanjie, Lv, Yisheng, Liu, Yu-Liang, Wang, Fei-Yue
Format: Article
Language:English
Subjects:
Algorithms
Deep learning
Detectors
Government agencies
Learning
Missing data
Neural networks
Traffic data imputation
Traffic engineering
Traffic flow
Transportation
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•The model denoising stacked autoencoders for traffic data imputation is proposed.•The model performance changes with temporal and spatial factors.•An efficient model realization with hierarchically training algorithm is developed. Traffic data provide the basis for both research and applications in transportation control, management, and evaluation, but real-world traffic data collected from loop detectors or other sensors often contain corrupted or missing data points which need to be imputed for traffic analysis. For this end, here we propose a deep learning model named denoising stacked autoencoders for traffic data imputation. We tested and evaluated the model performance with consideration of both temporal and spatial factors. Through these experiments and evaluation results, we developed an algorithm for efficient realization of deep learning for traffic data imputation by training the model hierarchically using the full set of data from all vehicle detector stations. Using data provided by Caltrans PeMS, we have shown that the mean absolute error of the proposed realization is under 10veh/5-min, a better performance compared with other popular models: the history model, ARIMA model and BP neural network model. We further investigated why the deep leaning model works well for traffic data imputation by visualizing the features extracted by the first hidden layer. Clearly, this work has demonstrated the effectiveness as well as efficiency of deep learning in the field of traffic data imputation and analysis.
ISSN:0968-090X
1879-2359
DOI:10.1016/j.trc.2016.09.015