ST-FVGAN: filling series traffic missing values with generative adversarial network
Authors:
- Bing YangBYBing Yang
- Yan KangYKYan Kang
- Yaoyao YuanYYYaoyao Yuan
- 2 February 2021
- journal article
- research article
- Published by Taylor & Francis in Transportation Letters
- Vol. 14 (4) , 407-415
Abstract
The imputation of time series traffic flow data is of great significance to the intelligent transportation, urban planning, and road emergency handling. This paper proposes a filling missing time series traffic data with Generative Adversarial Network (ST-FVGAN), which not only considers the spatio-temporal correlation and utilizes the idea of data generating of the Generative Adversarial Network, but also considers the external factors and introduces a more comprehensive loss function. Specifically, the model firstly constructs a Generator network which is composed of convolutional layer, residual block, and pixelshuffle block for the better potential distribution of the existing data, and then use the Discriminator network for the input judging. Experiments are conducted on the open-source TaxiBJ GPS dataset, and evaluated by the root mean square error (RMSE) index. The experimental results show that our model has the better accurate and reasonable performance than the traditional imputation methodsKeywords
Funding Information
- Code intelligent recommendation based on automatic neural architecture search under community evolution (2020SE303)
- Dynamic multi-objective requirement optimization based on transfer learning
This publication has 28 references indexed in Scilit:
- A domestic Adaptable Infant Monitoring System using wireless sensor networks2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- A functional data approach to missing value imputation and outlier detection for traffic flow dataTransportmetrica B: Transport Dynamics, 2014
- Efficient missing data imputing for traffic flow by considering temporal and spatial dependenceTransportation Research Part C: Emerging Technologies, 2013
- A tensor-based method for missing traffic data completionTransportation Research Part C: Emerging Technologies, 2013
- Non-parametric regression for space–time forecasting under missing dataComputers, Environment and Urban Systems, 2012
- Imputation of Missing Traffic Data during Holiday PeriodsTransportation Planning and Technology, 2008
- Matching Patterns for Updating Missing Values of Traffic CountsTransportation Planning and Technology, 2006
- Accurate freeway travel time prediction with state-space neural networks under missing dataTransportation Research Part C: Emerging Technologies, 2005
- Estimation of missing traffic counts using factor, genetic, neural, and regression techniquesTransportation Research Part C: Emerging Technologies, 2004