Fourier Graph Convolution Network for Time Series Prediction

Lyuchao Liao; Zhiyuan Hu; Chih-Yu Hsu; Jinya Su

doi:10.3390/math11071649

Fourier Graph Convolution Network for Time Series Prediction

Lyuchao Liao, Zhiyuan Hu, Chih-Yu Hsu^* (Corresponding Author), Jinya Su

^*Corresponding author for this work

Fujian University of Technology

Research output: Contribution to journal › Article › peer-review

3 Citations (Scopus)

5 Downloads (Pure)

Abstract

The spatio-temporal pattern recognition of time series data is critical to developing intelligent transportation systems. Traffic flow data are time series that exhibit patterns of periodicity and volatility. A novel robust Fourier Graph Convolution Network model is proposed to learn these patterns effectively. The model includes a Fourier Embedding module and a stackable Spatial-Temporal ChebyNet layer. The development of the Fourier Embedding module is based on the analysis of Fourier series theory and can capture periodicity features. The Spatial-Temporal ChebyNet layer is designed to model traffic flow’s volatility features for improving the system’s robustness. The Fourier Embedding module represents a periodic function with a Fourier series that can find the optimal coefficient and optimal frequency parameters. The Spatial-Temporal ChebyNet layer consists of a Fine-grained Volatility Module and a Temporal Volatility Module. Experiments in terms of prediction accuracy using two open datasets show the proposed model outperforms the state-of-the-art methods significantly.

Original language	English
Article number	1649
Journal	Mathematics
Volume	11
Issue number	7
DOIs	https://doi.org/10.3390/math11071649
Publication status	Published - 29 Mar 2023

Bibliographical note

This work was supported in part by the projects of the National Natural Science Foundation of China (41971340, 61304199), projects of Fujian Provincial Department of Science and Technology (2021Y4019, 2020D002, 2020L3014, 2019I0019, 2008Y3001), projects of Fujian Provincial Department of Finance (GY-Z230007), and the project of Fujian Provincial Universities Engineering Research Center for Intelligent Driving Technology (Fujian University of Technology) (KF-J21012).

Data Availability Statement

The original highway datasets are derived from [11], and further raw data used in this work can be obtained from the corresponding author upon request.

Keywords

traffic flow prediction
periodicity
volatility
Fourier embedding
spatial-temporal ChebyNet
graph convolutional neural network

Access to Document

10.3390/math11071649Licence: CC BY

Liao_etal_mathematics_Fourier_Graph_Convolution_VOR
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Final published version, 4.51 MBLicence: CC BY

Cite this

@article{b26d92c165f74713a682382972804f59,

title = "Fourier Graph Convolution Network for Time Series Prediction",

abstract = "The spatio-temporal pattern recognition of time series data is critical to developing intelligent transportation systems. Traffic flow data are time series that exhibit patterns of periodicity and volatility. A novel robust Fourier Graph Convolution Network model is proposed to learn these patterns effectively. The model includes a Fourier Embedding module and a stackable Spatial-Temporal ChebyNet layer. The development of the Fourier Embedding module is based on the analysis of Fourier series theory and can capture periodicity features. The Spatial-Temporal ChebyNet layer is designed to model traffic flow{\textquoteright}s volatility features for improving the system{\textquoteright}s robustness. The Fourier Embedding module represents a periodic function with a Fourier series that can find the optimal coefficient and optimal frequency parameters. The Spatial-Temporal ChebyNet layer consists of a Fine-grained Volatility Module and a Temporal Volatility Module. Experiments in terms of prediction accuracy using two open datasets show the proposed model outperforms the state-of-the-art methods significantly.",

keywords = "traffic flow prediction, periodicity, volatility, Fourier embedding, spatial-temporal ChebyNet, graph convolutional neural network",

author = "Lyuchao Liao and Zhiyuan Hu and Chih-Yu Hsu and Jinya Su",

note = "This work was supported in part by the projects of the National Natural Science Foundation of China (41971340, 61304199), projects of Fujian Provincial Department of Science and Technology (2021Y4019, 2020D002, 2020L3014, 2019I0019, 2008Y3001), projects of Fujian Provincial Department of Finance (GY-Z230007), and the project of Fujian Provincial Universities Engineering Research Center for Intelligent Driving Technology (Fujian University of Technology) (KF-J21012). ",

year = "2023",

month = mar,

day = "29",

doi = "10.3390/math11071649",

language = "English",

volume = "11",

journal = "Mathematics",

issn = "2227-7390",

publisher = "MDPI",

number = "7",

}

TY - JOUR

T1 - Fourier Graph Convolution Network for Time Series Prediction

AU - Liao, Lyuchao

AU - Hu, Zhiyuan

AU - Hsu, Chih-Yu

AU - Su, Jinya

N1 - This work was supported in part by the projects of the National Natural Science Foundation of China (41971340, 61304199), projects of Fujian Provincial Department of Science and Technology (2021Y4019, 2020D002, 2020L3014, 2019I0019, 2008Y3001), projects of Fujian Provincial Department of Finance (GY-Z230007), and the project of Fujian Provincial Universities Engineering Research Center for Intelligent Driving Technology (Fujian University of Technology) (KF-J21012).

PY - 2023/3/29

Y1 - 2023/3/29

N2 - The spatio-temporal pattern recognition of time series data is critical to developing intelligent transportation systems. Traffic flow data are time series that exhibit patterns of periodicity and volatility. A novel robust Fourier Graph Convolution Network model is proposed to learn these patterns effectively. The model includes a Fourier Embedding module and a stackable Spatial-Temporal ChebyNet layer. The development of the Fourier Embedding module is based on the analysis of Fourier series theory and can capture periodicity features. The Spatial-Temporal ChebyNet layer is designed to model traffic flow’s volatility features for improving the system’s robustness. The Fourier Embedding module represents a periodic function with a Fourier series that can find the optimal coefficient and optimal frequency parameters. The Spatial-Temporal ChebyNet layer consists of a Fine-grained Volatility Module and a Temporal Volatility Module. Experiments in terms of prediction accuracy using two open datasets show the proposed model outperforms the state-of-the-art methods significantly.

AB - The spatio-temporal pattern recognition of time series data is critical to developing intelligent transportation systems. Traffic flow data are time series that exhibit patterns of periodicity and volatility. A novel robust Fourier Graph Convolution Network model is proposed to learn these patterns effectively. The model includes a Fourier Embedding module and a stackable Spatial-Temporal ChebyNet layer. The development of the Fourier Embedding module is based on the analysis of Fourier series theory and can capture periodicity features. The Spatial-Temporal ChebyNet layer is designed to model traffic flow’s volatility features for improving the system’s robustness. The Fourier Embedding module represents a periodic function with a Fourier series that can find the optimal coefficient and optimal frequency parameters. The Spatial-Temporal ChebyNet layer consists of a Fine-grained Volatility Module and a Temporal Volatility Module. Experiments in terms of prediction accuracy using two open datasets show the proposed model outperforms the state-of-the-art methods significantly.

KW - traffic flow prediction

KW - periodicity

KW - volatility

KW - Fourier embedding

KW - spatial-temporal ChebyNet

KW - graph convolutional neural network

U2 - 10.3390/math11071649

DO - 10.3390/math11071649

M3 - Article

SN - 2227-7390

VL - 11

JO - Mathematics

JF - Mathematics

IS - 7

M1 - 1649

ER -

Fourier Graph Convolution Network for Time Series Prediction

Abstract

Bibliographical note

Data Availability Statement

Keywords

Access to Document

Fingerprint

Cite this