ST-FVGAN: filling series traffic missing values with generative adversarial network

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 methods

KEYWORDS:

• Generative adversarial networks (GAN)
• spatio-temporal features processing
• traffic data imputation
• urban computing

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the “Dynamic multi-objective requirement optimization based on transfer learning”,the National Science Foundation of China, Grant No. 61762092; “Code intelligent recommendation based on automatic neural architecture search under community evolution”, Key Laboratory in Software Engineering of Yunnan Province, Grant No. 2020SE303; “Genetic engineering of rare and precious metal materials in Yunnan Province”, Major projects of yunnan province, Grant No. 202002AB080001