Skip to main content

Advertisement

Multivariate time series classification based on spatial-temporal attention dynamic graph neural network

  • Published:
Applied Intelligence Aims and scope Submit manuscript

Abstract

Multivariate time series classification (MVTSC) has significant potential for Internet of Things applications. Recently, deep learning (DL) and graph neural network (GNN) methods have been applied to MVTSC tasks. Unfortunately, DL-based methods ignore explicit inter-series correlation modeling. Most existing GNN-based methods treat MVTS data as a static graph spanning the entire temporal trajectory, which inadequately captures changes in inter-series local correlations. To address this problem, we propose the spatial-temporal attention dynamic GNN (STADGNN), which explicitly models dynamic inter-series correlations by constructing the MVTS data into a dynamic graph structure at a finer granularity. It combines discrete Fourier transform (DFT) and discrete wavelet transform (DWT), which extract the global and local features of MVTS data in an end-to-end framework. In dynamic graph learning, spatial-temporal attention mechanisms are employed to simultaneously capture changes in inter-series local correlations and intra-series temporal dependencies without relying on predefined priors. Experimental results on 25 UEA datasets indicate that the STADGNN outperforms existing DL-based and GNN-based baseline models in MVTSC tasks.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

Data availability

Data openly available in a public repository. https://www.timeseriesclassification.com

Code availability

The available source code has been provided in the article.

Materials availability

Not applicable.

Notes

  1. https://www.timeseriesclassification.com

References

  1. Chandel A, Naruka MS (2023) Performance enhancement of smart grid integration using a novel intellectual multi-objective control technique. Soft Comput 27:12419–12437

    Article  MATH  Google Scholar 

  2. Doddaiah R, Parvatharaju PS, Rundensteiner E, Hartvigsen T (2024) Explaining deep multi-class time series classifiers. Knowl Inf Syst

  3. Li H, Yang S, Song Y, Luo Y, Li J, Zhou T (2022) Spatial dynamic graph convolutional network for traffic flow forecasting. Appl Intell 53:14986–14998

    Article  MATH  Google Scholar 

  4. Wu Y, Lian C, Zeng Z, Xu B, Su Y (2023) An aggregated convolutional transformer based on slices and channels for multivariate time series classification. IEEE Trans Emerg Top Comput Intell 7:768–779

  5. Zhou Q, Wang Z, Huang L (2023) Elm-kl-lstm: a robust and general incremental learning method for efficient classification of time series data. PeerJ Comput Sci 9

  6. Mei J, Liu M, Yf Wang, Gao H (2016) Learning a mahalanobis distance-based dynamic time warping measure for multivariate time series classification. IEEE Trans Cyberne 46:1363–1374

    Article  MATH  Google Scholar 

  7. Seto S, Zhang W, Zhou Y (2015) Multivariate time series classification using dynamic time warping template selection for human activity recognition. IEEE Symp Ser Comput Intell 1399–1406

  8. Zhang HB, Wang PL, Zhang MM, Wang W (2023) Shapelet based two-step time series positive and unlabeled learning. J Comput Sci Technol 38:1387–1402

    Article  MATH  Google Scholar 

  9. Chen J, Wan Y (2023) Localized shapelets selection for interpretable time series classification. Applied Intelligence 53:17985–18001

    Article  MATH  Google Scholar 

  10. Cai B, Huang GL, Yang S, Xiang Y, Chi CH (2023) Se-shapelets: Semi-supervised clustering of time series using representative shapelets. Expert Syst Appl 240:122584

    Article  Google Scholar 

  11. Dempster A, Schmidt DF, Webb GI (2022) Hydra: competing convolutional kernels for fast and accurate time series classification. Data Min Knowl Disc 37:1779–1805

    Article  MathSciNet  MATH  Google Scholar 

  12. Wang T, Liu Z, Zhang T, Hussain SF, Waqas MM, Li Y (2022) Adaptive feature fusion for time series classification. Knowl Based Syst 243:108459

    Article  MATH  Google Scholar 

  13. Yu Y, Si X, Hu C, Jx Zhang (2019) A review of recurrent neural networks: Lstm cells and network architectures. Neural Comput 31:1235–1270

    Article  MathSciNet  MATH  Google Scholar 

  14. Zhang X, Zhong C, Zhang J, Wang T, Ng WWY (2023) Robust recurrent neural networks for time series forecasting. Neurocomputing 526:143–157

    Article  MATH  Google Scholar 

  15. Ahmed S, Nielsen IE, Tripathi A, Siddiqui S, Ramachandran RP, Rasool G (2022) Transformers in time-series analysis: A tutorial. Circuits, Systems, and Signal Processing 42:7433–7466

    Article  MATH  Google Scholar 

  16. Khan A, Rauf Z, Sohail A, Rehman A, Asif H, Asif A, Farooq U (2023) A survey of the vision transformers and their cnn-transformer based variants. Artif Intell Rev 56:2917–2970

    Article  MATH  Google Scholar 

  17. Zhu Y, Luo S, Huang D, Zheng W, Su F, Hou B (2023) Drcnn: decomposing residual convolutional neural networks for time series forecasting. Sci Rep 13

  18. Yang Y, Lu J (2022) Foreformer: an enhanced transformer-based framework for multivariate time series forecasting. Appl Intell 53:12521–12540

    Article  MATH  Google Scholar 

  19. Xiao Z, Xu X, Zhang H, Szczerbicki E (2021) A new multi-process collaborative architecture for time series classification. Knowl Based Syst 220:106934

    Article  MATH  Google Scholar 

  20. Guo S, Lin Y, Wan H, Li X, Cong G (2021) Learning dynamics and heterogeneity of spatial-temporal graph data for traffic forecasting. IEEE Trans Knowl Data Eng 34:5415–5428

    Article  MATH  Google Scholar 

  21. Huang B, Dou H, Luo Y, Li J, Wang J, Zhou T (2023) Adaptive spatiotemporal transformer graph network for traffic flow forecasting by iot loop detectors. IEEE Int Things J 10:1642–1653

    Article  Google Scholar 

  22. Jin G, Li F, Zhang J, Wang MF, Huang J (2022) Automated dilated spatio-temporal synchronous graph modeling for traffic prediction. IEEE Trans Intell Trans Syst 24:8820–8830

    Article  MATH  Google Scholar 

  23. Jin M, Zheng Y, Li Y, Chen S, Yang B, Pan S (2022) Multivariate time series forecasting with dynamic graph neural odes. IEEE Trans Knowl Data Eng 35:9168–9180

    Article  MATH  Google Scholar 

  24. Cini A, Zambon D, Alippi C (2022) Sparse graph learning from spatiotemporal time series. J Mach Learn Res 24:242–124236

    MathSciNet  MATH  Google Scholar 

  25. Chen L, Chen D, Shang Z, Wu B, Zheng C, Wen B, Zhang W (2022) Multi-scale adaptive graph neural network for multivariate time series forecasting. IEEE Trans Knowl Data Eng 35:10748–10761

    Article  MATH  Google Scholar 

  26. Zheng Y, Koh HY, Jin M, Chi L, Phan KT, Pan S, Chen YPP, Xiang W (2023) Correlation-aware spatial-temporal graph learning for multivariate time-series anomaly detection. IEEE Trans Neural Netw Learn Syst PP

  27. Zhang K, Wen Q, Zhang C, Cai R, Jin M, Liu Y, Zhang J, Liang Y, Pang G, Song D, Pan S (2023) Self-supervised learning for time series analysis: Taxonomy, progress, and prospects. IEEE Trans Patt Anal Mach Intell PP

  28. Lyu X, Chen Y, Chen J, Liu X, Gao J (2022) Spatio-temporal based architecture topology search for multivariate time series prediction. IEEE International Conference on Big Data (Big Data) 2022:1304–1309

    Article  MATH  Google Scholar 

  29. Li G, Choi B, Xu J, Bhowmick SS, Chun KP, Wong GLH (2021) Efficient shapelet discovery for time series classification (extended abstract). 2021 IEEE 37th International Conference on Data Engineering (ICDE), 2336–2337

  30. Yang J, Jing S, Huang G (2022) Accurate and fast time series classification based on compressed random shapelet forest. Appl Intell 53:5240–5258

    MATH  Google Scholar 

  31. Ji C, Zhao C, Liu S, Yang C, Pan L, Wu L, Meng X (2019) A fast shapelet selection algorithm for time series classification. Comput. Netw 148:231–240

    Article  MATH  Google Scholar 

  32. Yuan J, Shi M, Wang Z, Liu H, Li J (2022) Random pairwise shapelets forest: an effective classifier for time series. Knowl Inf Syst 64:143–174

    Article  MATH  Google Scholar 

  33. Li F, Feng J, Yan H, Jin G, Jin D, Li Y (2021) Dynamic graph convolutional recurrent network for traffic prediction: Benchmark and solution. ACM Trans Knowl Discov Data 17:1–21

    MATH  Google Scholar 

  34. Ruiz AP, Flynn M, Large J, Middlehurst M, Bagnall A (2020) The great multivariate time series classification bake off: a review and experimental evaluation of recent algorithmic advances. J Data Min Knowl Discov 35:401–449

    Article  MathSciNet  MATH  Google Scholar 

  35. Zheng Z, Zhou M, Chen Y, Huo M, Sun L, Zhao S, Chen D (2022) A fused method of machine learning and dynamic time warping for road anomalies detection. IEEE Trans Intell Transp Syst 23:827–839

    Article  MATH  Google Scholar 

  36. Belkhouja T, Yan Y, Doppa JR (2022) Dynamic time warping based adversarial framework for time-series domain. IEEE Trans Patt Anal Mach Intell 45:7353–7366

    Article  MATH  Google Scholar 

  37. Teng Y, Wang G, He C, Wu Y, Li C (2023) Optimization of dynamic time warping algorithm for abnormal signal detection. Int J Data Sci Anal 1–13

  38. Peng B, Ding Y, Xia Q, Yang Y (2023) Recurrent neural networks integrate multiple graph operators for spatial time series prediction. Appl Intell 53:26067–26078

    Article  MATH  Google Scholar 

  39. Jordan I, Sokól PA, Park IM (2019) Gated recurrent units viewed through the lens of continuous time dynamical systems. Front Comput Neurosci 15

  40. Liu X, Luo Y, Ye Y, Lu J (2017) Mc-dcnn: Dilated convolutional neural network for computing stereo matching cost. In: International Conference on Neural Information Processing. https://api.semanticscholar.org/CorpusID:28982234

  41. Liu J, Liu Y, Li D, Wang H, Huang X, Song L (2023) Dsdcla: driving style detection via hybrid cnn-lstm with multi-level attention fusion. Appl Intell 1–18

  42. Liu F, Zhou X, Cao J, Wang Z, Wang T, Wang H, Zhang Y (2022) Anomaly detection in quasi-periodic time series based on automatic data segmentation and attentional lstm-cnn. IEEE Trans Knowl Data Eng 34:2626–2640

    Article  MATH  Google Scholar 

  43. Tang W, Long G, Liu L, Zhou T, Blumenstein M, Jiang J (2020) Omni-scale cnns: a simple and effective kernel size configuration for time series classification. In: International Conference on Learning Representations. https://api.semanticscholar.org/CorpusID:249848305

  44. Khan SH, Naseer M, Hayat M, Zamir SW, Khan FS, Shah M (2021) Transformers in vision: A survey. ACM Comput Surv (CSUR) 54:1–41

    Article  MATH  Google Scholar 

  45. Wang W, Yao L, Chen L, Lin B, Cai D, He X, Liu W (2021) Crossformer: A versatile vision transformer hinging on cross-scale attention. In: International Conference on Learning Representations. https://api.semanticscholar.org/CorpusID:238531695

  46. Gupta A, Maurya MK, Goyal N, Chaurasiya VK (2023) Istgcn: Integrated spatio-temporal modeling for traffic prediction using traffic graph convolution network. Appl Intell 53:29153–29168

    Article  MATH  Google Scholar 

  47. Wu H, Xu J, Wang J, Long M (2021) Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting. In: Neural Information Processing Systems. https://api.semanticscholar.org/CorpusID:235623791

  48. Wang Y, Zhu J, Kang R (2023) Destformer: A transformer based on explicit seasonal–trend decomposition for long-term series forecasting. Appl Sci

  49. Wu Z, Pan S, Long G, Jiang J, Chang X, Zhang C (2020) Connecting the dots: Multivariate time series forecasting with graph neural networks. Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining

  50. Spadon G, Qiao Brandoli B, Matwin S, Rodrigues-Jr JF, Sun J (2020) Pay attention to evolution: Time series forecasting with deep graph-evolution learning. IEEE Trans Patt Anal Mach Intell 44:5368–5384

    Article  Google Scholar 

  51. Wang X, Wang Y, Peng J, Zhang Z (2023) Multivariate long sequence time-series forecasting using dynamic graph learning. J Ambient Intell Human Comput 14:7679–7693

    Article  MATH  Google Scholar 

  52. Deng J, Chen X, Jiang R, Yin D, Yang Y, Song X, Tsang IWH (2023) Disentangling structured components: Towards adaptive, interpretable and scalable time series forecasting. IEEE Trans Knowl Data Eng

  53. Wang Y, Xu Y, Yang J, Wu M, Li X, Xie L, Chen Z (2020) Fully-connected spatial-temporal graph for multivariate time series data. In: AAAI Conference on Artificial Intelligence. https://api.semanticscholar.org/CorpusID:261682449

  54. Kuo PC, Chou YT, Li KY, Chang WT, Huang YN, Chen CS (2024) Gnn-lstm-based fusion model for structural dynamic responses prediction. Eng Struct

  55. Yang X, Li H, Feng X, Jin Z () Variable-wise generative adversarial transformer in multivariate time series anomaly detection. Appl Intell 53:28745–28767

  56. Zhang Z, Zhao Y, Liu J, Wang S, Tao R, Xin R, Zhang J (2018) A general deep learning framework for network reconstruction and dynamics learning. Appl Netw Sci 4

  57. Maaten L, Hinton GE (2008) Visualizing data using t-sne. J Mach Learn Res 9:2579–2605

    MATH  Google Scholar 

  58. Wu Y, Lian C, Zeng Z, Xu B, Su Y (2023) An aggregated convolutional transformer based on slices and channels for multivariate time series classification. IEEE Trans Emerg Top Comput Intell 7:768–779

    Article  MATH  Google Scholar 

Download references

Acknowledgements

The work was supported by the State Grid Corporation Science and Technology Project: Application research of a new generation of artificial intelligence in the security analysis and decision-making of future power grid (5100-202099522A-0-0-00).

Funding

Not applicable.

Author information

Authors and Affiliations

Authors

Contributions

Qiong Zuo contributed to the conception of the study; Lipeng Qian performed the experiment and wrote the manuscript; Haiguang Liu contributed significantly to analysis; Hong Zhu helped perform the analysis with constructive discussions.

Corresponding author

Correspondence to Qiong Zuo.

Ethics declarations

Conflict of interest/Competing interests

The authors have no competing interests to declare that are relevant to the content of this article.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Qian, L., Zuo, Q., Liu, H. et al. Multivariate time series classification based on spatial-temporal attention dynamic graph neural network. Appl Intell 55, 115 (2025). https://doi.org/10.1007/s10489-024-06014-8

Download citation

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s10489-024-06014-8

Keywords