Fourier Graph Convolution Network for Time Series Prediction
Abstract
:1. Introduction
 The existing methods learn the periodicity based on frequencydomain methods, such as spectral analysis and traditional Fourier Transform [14,15,16,17]. These models generally require manual parameters and comply with rigorous assumptions, making these methods incapable of capturing various periodicities.
 There is still a lack of an efficient way to learn dynamic volatility for improving robusticity, which is crucial to the dynamic spatialtemporal pattern recognition of the traffic network.
 Some models capture periodicity and volatility, but these methods capture them independently and ignore their inherent relationship.
 A novel Fourier Embedding module is proposed to capture periodicity patterns, which is proven to learn diversified periodicity patterns.
 A stackable SpatialTemporal ChebyNet layer, including a Finegrained Volatility Module and a Temporal Volatility Module, is proposed to handle the complex volatility and learn dynamic temporal volatility for improving the system’s robusticity.
 A dynamic Fourier Graph Convolution Network framework is proposed to integrate the periodicity and volatility analysis, which could be easily trained in an endtoend method. Extensive experiments are conducted on several realworld traffic flow data, and the results significantly outperform stateoftheart methods.
2. Literature Review
2.1. Traffic Flow Data Decomposition
2.2. Graph Convolution Network
3. Methods
3.1. Preliminaries
3.1.1. The Complex Fourier Series
3.1.2. Real Fourier Series
3.1.3. Problem Statement
3.2. Fourier Graph Convolution Network
3.2.1. Data Construction
3.2.2. Fourier Embedding
3.2.3. SpatialTemporal ChebyNet Layer
 A.
 Finegrained Volatility Module
 B.
 Temporal Volatility Module
3.3. Fusion & Loss Function
Algorithm 1 Pseudocode for the FGCN model 
Input: $\u2460$ The FGCN input feature ${\mathcal{X}}_{\mathcal{G}}\in {\mathbb{R}}^{F\times N\times T}$, including the week period ${\mathcal{X}}_{\mathcal{G}}^{{\mathit{T}}_{\mathit{w}}}\in {\mathbb{R}}^{F\times N\times {T}_{w}}$, day period ${\mathcal{X}}_{\mathcal{G}}^{{\mathit{T}}_{\mathit{d}}}\in {\mathbb{R}}^{F\times N\times {T}_{d}}$, and recentperiod ${\mathcal{X}}_{\mathcal{G}}^{{\mathit{T}}_{\mathit{r}}}\in {\mathbb{R}}^{F\times N\times {T}_{r}}$; $\u2461$ Laplacian matrix $L\in {\mathbb{R}}^{N\times N}$; Output: $\widehat{Y}\in {\mathbb{R}}^{C\times N\times T}$

4. Results and the Discussion
4.1. Data Description
4.2. Evaluation Metrics
4.3. Experimental Settings
4.4. Baselines and StateoftheArt Methods
4.5. Experiment Results
4.6. Performance of FE and STCN Modules
5. Conclusions
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
Nomenclatures
AGCRN  Adaptive Graph Convolutional Recurrent Network 
ASTGCN  Attentionbased SpatialTemporal Graph Convolution Network 
ARCH  Autoregressive Conditional Heteroskedasticity 
ARIMA  Autoregressive Integrated Moving Average 
BV  Boundedly Varied 
CFS  Complex Fourier Series 
CNN  Convolution Neural Networks 
DC  Dirichlet Condition 
DGCN  Dynamic Graph Convolution Network 
EMD  Empirical Mode Decomposition 
EEMD  Ensemble Empirical Mode Decomposition 
ELM  Extreme Learning Machine 
FE  Fourier Embedding 
FGCN  Fourier Graph Convolution Network 
GRU  Gated Recurrent Unit networks 
GCN  Graph Convolution Network 
HA  Historical Average 
iGCGCN  improved Dynamic Chebyshev Graph Convolution Network 
ITS  Intelligent Transportation Systems 
MAE  Mean Absolute Error 
MAPE  Mean Absolute Percentage Error 
MSE  Mean Square Error 
PeMS  Performance Measurement System 
RFS  Real Fourier series 
RMSE  Root Mean Square Error 
SP  Signal Processing 
STCN  SpatialTemporal ChebyNet 
STSGCN  SpatialTemporal Synchronous Graph Convolutional Networks 
STGCN  SpatioTemporal Graph Convolutional Network 
SVR  Support Vector Regression 
TGCN  Temporal Graph Convolutional Network 
TSASL  TimeSeries Analysis and SupervisedLearning 
$\left\xb7\right$  The length of a set. 
$\odot $  Hadamard product. 
[△, ○]  The concatenation of $\u25b3$ and $\u25cb$. 
σ(∙)  The sigmoid function. 
$\mathrm{sin}\left(\xb7\right)$, $\mathrm{cos}\left(\xb7\right)$  The sine and cosine functions. 
⋆  The Causal convolution operator. 
∗  The Graph convolution operator. 
$\mathcal{G}$  A graph. 
V  The set of nodes in a graph. 
E  The set of edges in a graph. 
A  The adjacency matrix of the graph. 
D  The degree matrix of $A$, $\mathrm{D}={{\displaystyle \sum}}_{j=0}^{N}{A}_{i,j}$. 
L  The Laplacian matrix $\mathit{L}=DA$. 
${e}_{i,j}$  The edge between node $i$ and node $j$. 
${\mathcal{X}}_{\mathcal{G}}\in {\mathbb{R}}^{F\times N\times T}$  The spatialtemporal graph by Data Construction. 
${X}_{\mathcal{G}}\in {\mathbb{R}}^{F\times N}$  The spatialtemporal graph in the case of $d=1$. 
${\mathcal{X}}_{\mathcal{G}}^{\u2033}\in {\mathbb{R}}^{F\times N\times T}$  The output of the FE module. 
${\mathcal{X}}_{\mathcal{G}}^{res}\in {\mathbb{R}}^{F\times N\times T}$  The residual ${\mathcal{X}}_{\mathcal{G}}^{\mathrm{res}}={\mathcal{X}}_{\mathcal{G}}^{\u2033}+{\mathcal{X}}_{\mathcal{G}}$. 
${\mathcal{X}}_{\mathit{g}}\in {\mathbb{R}}^{C\times N\times T}$  The output of the Finegrained Volatility Module. 
${\mathcal{X}}_{\mathit{t}\mathit{e}\mathit{m}\mathit{p}}\in {\mathbb{R}}^{C\times N\times T}$  The output of the Temporal Volatility Module. 
$F$  The number of original characteristics. 
$N$  The nodes of the graph. 
$T$  The number of time slices of the graph. 
$d$  The length of the vector embedding. 
${T}_{h},{T}_{f}$  The length of historical data and prediction data. 
$M$  The order of the Fourier polynomial in the FE module. 
References
 Duan, H.; Wang, G. Partial differential grey model based on control matrix and its application in shortterm traffic flow prediction. Appl. Math. Model. 2023, 116, 763–785. [Google Scholar] [CrossRef]
 Han, Y.; Zhao, S.; Deng, H.; Jia, W. Principal graph embedding convolutional recurrent network for traffic flow prediction. Appl. Intell. 2023, 1–15. [Google Scholar] [CrossRef]
 Guo, K.; Hu, Y.; Qian, Z.; Sun, Y.; Gao, J.; Yin, B. Dynamic graph convolution network for traffic forecasting based on latent network of laplace matrix estimation. IEEE Trans. Intell. Transp. Syst. 2020, 23, 1009–1018. [Google Scholar] [CrossRef]
 Li, W.; Wang, X.; Zhang, Y.; Wu, Q. Traffic flow prediction over mutisensor data correlation with graph convolution network. Neurocomputing 2021, 427, 50–63. [Google Scholar] [CrossRef]
 Xue, Y.; Tang, Y.; Xu, X.; Liang, J.; Neri, F. Multiobjective feature selection with missing data in classification. IEEE Trans. Emerg. Top. Comput. Intell. 2021, 6, 355–364. [Google Scholar] [CrossRef]
 Williams, B.M.; Hoel, L.A. Modeling and forecasting vehicular traffic flow as a seasonal ARIMA process: Theoretical basis and empirical results. J. Transp. Eng. 2003, 129, 664–672. [Google Scholar] [CrossRef] [Green Version]
 Engle, R. Risk and volatility: Econometric models and financial practice. Am. Econ. Rev. 2004, 94, 405–420. [Google Scholar] [CrossRef] [Green Version]
 Xue, Y.; Xue, B.; Zhang, M. Selfadaptive particle swarm optimization for largescale feature selection in classification. ACM Trans. Knowl. Discov. Data 2019, 13, 1–27. [Google Scholar] [CrossRef]
 Xue, Y.; Wang, Y.; Liang, J.; Slowik, A. A selfadaptive mutation neural architecture search algorithm based on blocks. IEEE Comput. Intell. Mag. 2021, 16, 67–78. [Google Scholar] [CrossRef]
 Zhao, L.; Song, Y.; Zhang, C.; Liu, Y.; Wang, P.; Lin, T.; Deng, M.; Li, H. TGCN: A Temporal Graph Convolutional Network for Traffic Prediction. IEEE Trans. Intell. Transp. Syst. 2019, 21, 3848–3858. [Google Scholar] [CrossRef] [Green Version]
 Guo, S.; Lin, Y.; Feng, N.; Song, C.; Wan, H. Attention based spatialtemporal graph convolutional networks for traffic flow forecasting. In Proceedings of the AAAI’19: AAAI Conference on Artificial Intelligence, Honolulu, HI, USA, 27 January–1 February 2019. [Google Scholar] [CrossRef] [Green Version]
 Song, C.; Lin, Y.; Guo, S.; Wan, H. SpatialTemporal Synchronous Graph Convolutional Networks: A New Framework for SpatialTemporal Network Data Forecasting. In Proceedings of the AAAI Conference on Artificial Intelligence, New York, NY, USA, 7–12 February 2020. [Google Scholar] [CrossRef]
 Liao, L.; Hu, Z.; Zheng, Y.; Bi, S.; Zou, F.; Qiu, H.; Zhang, M. An improved dynamic Chebyshev graph convolution network for traffic flow prediction with spatialtemporal attention. Appl. Intell. 2022, 52, 16104–16116. [Google Scholar] [CrossRef]
 Chen, L.; Zheng, L.; Yang, J.; Xia, D.; Liu, W. Shortterm traffic flow prediction: From the perspective of traffic flow decomposition. Neurocomputing 2020, 413, 444–456. [Google Scholar] [CrossRef]
 Zhang, Y.; Zhang, Y.; Haghani, A. A hybrid shortterm traffic flow forecasting method based on spectral analysis and statistical volatility model. Transp. Res. Part C Emerg. Technol. 2014, 43, 65–78. [Google Scholar] [CrossRef]
 Chen, X.; Lu, J.; Zhao, J.; Qu, Z.; Yang, Y.; Xian, J. Traffic flow prediction at varied time scales via ensemble empirical mode decomposition and artificial neural network. Sustainability 2020, 12, 3678. [Google Scholar] [CrossRef]
 Tian, Z. Approach for shortterm traffic flow prediction based on empirical mode decomposition and combination model fusion. IEEE Trans. Intell. Transp. Syst. 2020, 22, 5566–5576. [Google Scholar] [CrossRef]
 Zivot, E.; Wang, J. Vector autoregressive models for multivariate time series. Model. Financ. Time Ser. SPlus^{®} 2006, 385–429. [Google Scholar] [CrossRef]
 Sun, S.; Zhang, C.; Yu, G. A bayesian network approach to traffic flow forecasting. IEEE Trans. Intell. Transp. Syst. 2006, 7, 124–132. [Google Scholar] [CrossRef]
 Liu, L.; Zhen, J.; Li, G.; Zhan, G.; He, Z.; Du, B.; Lin, L. Dynamic spatialtemporal representation learning for traffic flow prediction. IEEE Trans. Intell. Transp. Syst. 2020, 22, 7169–7183. [Google Scholar] [CrossRef]
 Shu, W.; Cai, K.; Xiong, N.N. A ShortTerm Traffic Flow Prediction Model Based on an Improved Gate Recurrent Unit Neural Network. IEEE Trans. Intell. Transp. Syst. 2021, 23, 16654–16665. [Google Scholar] [CrossRef]
 Guo, S.; Lin, Y.; Li, S.; Chen, Z.; Wan, H. Deep spatial–temporal 3D convolutional neural networks for traffic data forecasting. IEEE Trans. Intell. Transp. Syst. 2019, 20, 3913–3926. [Google Scholar] [CrossRef]
 Bruna, J.; Zaremba, W.; Szlam, A.; LeCun, Y. Spectral networks and locally connected networks on graphs. arXiv 2013, arXiv:1312.6203. [Google Scholar] [CrossRef]
 Defferrard, M.; Bresson, X.; Vandergheynst, P. Convolutional neural networks on graphs with fast localized spectral filtering. Adv. Neural Inf. Process. Syst. 2016, 29. [Google Scholar] [CrossRef]
 Kipf, T.N.; Welling, M. Semisupervised classification with graph convolutional networks. arXiv 2016, arXiv:1609.02907. [Google Scholar] [CrossRef]
 Kearnes, S.; McCloskey, K.; Berndl, M.; Pande, V.; Riley, P. Molecular graph convolutions: Moving beyond fingerprints. J. Comput.Aided Mol. Des. 2016, 30, 595–608. [Google Scholar] [CrossRef] [Green Version]
 Niepert, M.; Ahmed, M.; Kutzkov, K. Learning convolutional neural networks for graphs. arXiv 2016, arXiv:1605.05273. [Google Scholar] [CrossRef]
 Hamilton, W.L.; Ying, R.; Leskovec, J. Inductive representation learning on large graphs. Adv. Neural Inf. Process. Syst. 2017, 30. [Google Scholar] [CrossRef]
 Chen, C.; Li, K.; Teo, S.G.; Zou, X.; Wang, K.; Wang, J.; Zeng, Z. Gated residual recurrent graph neural networks for traffic prediction. In Proceedings of the AAAI Conference on Artificial Intelligence, Honolulu, HI, USA, 27 January–1 February 2019. [Google Scholar] [CrossRef] [Green Version]
 Bai, L.; Yao, L.; Li, C.; Wang, X.; Wang, C. Adaptive graph convolutional recurrent network for traffic forecasting. arXiv 2020, arXiv:2007.02842. [Google Scholar] [CrossRef]
 Kazemi, S.M.; Goel, R.; Eghbali, S.; Ramanan, J.; Sahota, J.; Thakur, S.; Wu, S.; Smyth, C.; Poupart, P.; Brubaker, M. Time2vec: Learning a vector representation of time. arXiv 2019, arXiv:1907.05321. [Google Scholar] [CrossRef]
 Meyer, F.; van der Merwe, B.; Coetsee, D.J.J.U.C.S. Learning Concept Embeddings from Temporal Data. J. Univers. Comput. Sci. 2018, 24, 1378–1402. [Google Scholar] [CrossRef]
 Hsu, C.Y.; Huang, H.Y.; Lee, L.T. An interactive procedure to preserve the desired edges during the image processing of noise reduction. EURASIP J. Adv. Signal Process. 2010, 2010, 923748. [Google Scholar] [CrossRef] [Green Version]
 He, K.; Zhang, X.; Ren, S.; Sun, J. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016. [Google Scholar] [CrossRef]
 Vaswani, A.; Shazeer, N.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A.N.; Kaiser, Ł.; Polosukhin, I. Attention is all you need. Adv. Neural Inf. Process. Syst. 2017, 30. [Google Scholar] [CrossRef]
 Xue, Y.; Qin, J. Partial connection based on channel attention for differentiable neural architecture search. IEEE Trans. Ind. Inform. 2022, 1–10. [Google Scholar] [CrossRef]
 Chen, C.; Petty, K.; Skabardonis, A.; Varaiya, P.; Jia, Z. Freeway performance measurement system: Mining loop detector data. Transp. Res. Rec. 2001, 1748, 96–102. [Google Scholar] [CrossRef] [Green Version]
Models  HA  ARIMA  GRU  STGCN  ASTGCN  STSGCN  AGCRN  FGCN  

Datasets  Time  Metrics  
PeMSD4  15 min  MAE  30.505  28.366  24.240  21.191  20.448  20.019  18.850  18.358 
RMSE  41.873  36.443  36.458  33.235  32.072  31.927  30.970  29.429  
MAPE  27.043  47.375  19.561  18.921  15.210  13.530  12.536  12.956  
30 min  MAE  37.245  34.455  25.732  23.909  20.735  21.543  19.520  18.844  
RMSE  50.054  45.595  38.671  35.743  32.780  34.180  32.130  30.292  
MAPE  36.648  50.316  20.879  20.465  15.146  14.320  12.962  13.166  
45 min  MAE  43.930  42.042  27.433  25.727  21.048  23.053  20.040  19.244  
RMSE  58.101  59.311  41.149  38.362  33.453  36.390  33.100  30.975  
MAPE(%)  53.502  51.071  22.553  22.193  15.216  15.260  13.310  13.390  
60 min  MAE  50.539  52.997  29.408  27.617  21.494  24.627  20.960  19.603  
RMSE  65.982  77.380  44.017  41.077  34.247  38.563  34.420  31.555  
MAPE(%)  72.040  54.230  24.701  24.054  15.500  16.410  13.889  13.569  
PeMSD8  15 min  MAE  25.157  32.571  19.206  17.542  16.779  16.599  15.080  13.646 
RMSE  34.234  34.120  29.764  25.871  24.941  25.371  23.730  21.384  
MAPE(%)  16.053  22.634  13.629  13.080  11.888  10.989  9.650  9.424  
30 min  MAE  30.945  38.310  20.452  18.774  17.069  17.849  16.090  14.013  
RMSE  41.130  43.402  31.687  28.038  25.600  27.280  25.570  22.171  
MAPE(%)  20.438  30.260  15.048  13.917  11.842  11.566  10.183  9.634  
45 min  MAE  36.689  42.830  21.928  20.040  17.387  18.903  16.960  14.269  
RMSE  47.836  47.158  33.818  30.150  26.257  28.933  26.950  22.742  
MAPE(%)  25.163  35.444  16.799  14.867  11.933  12.200  10.736  9.792  
60 min  MAE  42.364  42.860  23.675  21.362  17.874  20.116  18.170  14.516  
RMSE  54.379  45.810  36.333  32.223  27.088  30.642  28.710  23.230  
MAPE(%)  30.236  35.495  18.986  15.923  12.210  13.040  11.514  9.924 
Dataset  Order  MAE  RMSE  MAPE  s/Epoch 

PeMSD8  2  14.14  22.29  9.82  52.59 
3  14.02  22.20  9.54  56.71  
4  14.29  22.42  9.76  59.92  
5  14.03  22.26  9.69  66.62  
PeMSD4  2  18.98  30.36  14.27  93.06 
3  18.86  30.29  13.21  104.17  
4  20.42  32.49  14.79  107.97  
5  19.60  31.11  13.63  116.93 
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. 
© 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/).
Share and Cite
Liao, L.; Hu, Z.; Hsu, C.Y.; Su, J. Fourier Graph Convolution Network for Time Series Prediction. Mathematics 2023, 11, 1649. https://doi.org/10.3390/math11071649
Liao L, Hu Z, Hsu CY, Su J. Fourier Graph Convolution Network for Time Series Prediction. Mathematics. 2023; 11(7):1649. https://doi.org/10.3390/math11071649
Chicago/Turabian StyleLiao, Lyuchao, Zhiyuan Hu, ChihYu Hsu, and Jinya Su. 2023. "Fourier Graph Convolution Network for Time Series Prediction" Mathematics 11, no. 7: 1649. https://doi.org/10.3390/math11071649