Abstract
Spatiotemporal modelling of short-term forecasts of metro passenger flows continue to face tremendous challenges. First, there is a need to consider the functional domain made up of several similar stations; Secondly, complex spatiotemporal models depend on a large number of learnable parameters. This paper proposed a spatiotemporal dual self-adaptive network based on the cluster (CG-TaLK) to accurately predict the inflow and outflow of subway passengers. Specifically, through the division of clustering, the members of each group learn a shared embedding, and use the inner product of embedding to mine the flow pattern between urban functional areas, so as to provide more accurate spatial information for prediction. In addition, in order to limit the number of parameters, we migrate a temporal adaptive convolution (TaLK) to capture the time correlation of each station according to the characteristics of passenger flow. The self-adaptive mechanism in space and time can enhance the fitting ability of the model. By comparing six representative algorithms on Hangzhou Metro dataset, the results show that the proposed method is effective and takes up the least parameters. Meanwhile, experiments show that the algorithm can find the main communication between function areas.
Similar content being viewed by others
References
Tascikaraoglu A (2018) Evaluation of spatio-temporal forecasting methods in various smart city applications. Renew Sust Energ Rev 82:424–435. https://doi.org/10.1016/j.rser.2017.09.078
Lee H, Rhee W (2019) DDP-GCN: Multi-graph convolutional network for spatiotemporal traffic forecasting. https://arxiv.org/abs/1905.12256
Lu H, Huang D, Song Y, Jiang D, Zhou T, Qin J (2020) ST-Trafficnet: a spatial-temporal deep learning network for traffic forecasting. Electronics 9(1474):1–17. https://doi.org/10.3390/electronics9091474
Hamed M, Al-Masaeid H, Said Z (1995) Short-term prediction of traffic volume in urban arterials. J Trans Eng 121(3):249–254. https://doi.org/10.1061/(ASCE)0733-947X(1995)121:3(249)
Zhao Y, Wu J, Zhang Y, Zhou B (2012) Wavelet-based kalman filter for traffic flow forecasting in sensornets. Inf Technol J 11(10):1518–1522. https://doi.org/10.3923/itj.2012.1518.1522
Zhao Y, Liu Y, Shan L, Zhou B (2012) Dynamic analysis of kalman filter for traffic flow forecasting in sensornets. Inf Technol J 11(10):1508–1512. https://doi.org/10.3923/itj.2012.1508.1512
Xia Z, Xue S, Wu L, Sun J, Chen Y, Zhang R (2020) ForeXGBoost: Passenger car sales prediction based on xgboost. Distributed and Parallel Databases 38(3):713–738. https://doi.org/10.1007/s10619-020-07294-y
Wang Y, Chen J, Chen J, Zeng X, Kong Y, Sun S, Guo Y, Liu Y (2020) Short-term load forecasting for industrial customers based on tcn-lightgbm. IEEE Trans Power Syst 3:1–14. https://doi.org/10.1109/TPWRS.2020.3028133
Doubravová J, Wiszniowski J, Horálek J (2016) Single layer recurrent neural network for detection of swarm-like earthquakes in W-Bohemia/Vogtlandthe-method. Comput Geosci 93:138–149. https://doi.org/10.1016/j.cageo.2016.05.011
Rather A, Agarwal A, Sastry V (2015) Recurrent neural network and a hybrid model for prediction of stock returns. Expert Syst Appl 42(6):3234–3241. https://doi.org/10.1016/j.eswa.2014.12.003
Zhao Z, Chen W, Wu X, Chen P, Liu J (2017) LSTM Network:a deep learning approach for short-term traffic forecast. IET Intell Transp Syst 11(1):68–75. https://doi.org/10.1049/iet-its.2016.0208
Xu Y, Han Y, Hong R, Tian Q (2018) Sequential video VLAD: Training the aggregation locally and temporally. IEEE Trans Image Process 27(10):4933–4944. https://doi.org/10.1109/TIP.2018.2846664
Li Y, Yu R, Shahabi C, Liu Y (2018) Diffusion convolutional recurrent neural network: Data-driven traffic forecasting. In: Proceedings of the 6th international conference on learning representations (ICLR-18). arXiv:1707.01926
Xu M, Dai W, Liu C, Gao X, Lin W, Qi G, Xiong H (2020) Spatial-temporal transformer networks for traffic flow forecasting. arXiv:2001.02908v1
Wang S, Li Q, Zhao C, Zhu X, Yuan H, Dai T (2021) Extreme clustering - a clustering method via density extreme points. Inf Sci 542:24–39. https://doi.org/10.1016/j.ins.2020.06.069
Fu T (2011) A review on time series data mining. Eng Appl of Artif Intell 24(1):164–181. https://doi.org/10.1016/j.engappai.2010.09.007
Niennattrakul V, Ratanamahatana C (2007) On clustering multimedia time series data using k-means and dynamic time warping. In: Proceedings of the 1st international conference on multimedia and ubiquitous rngineering (MUE-2007), pp 733–738. https://doi.org/10.1109/MUE.2007.165
Battke F, Symons S, Nieselt K (2010) Mayday-integrative analytics for expression data. BMC Bioinforma 11(121):1–10. https://doi.org/10.1186/1471-2105-11-121
Bruna J, Zaremba W, Szlam A, LeCun Y (2014) Spectral networks and locally connected networks on graphs. In: Proceedings of the 2th international conference on learning representations (ICLR-14). arXiv:1312.6203
Defferrard M, Bresson X, Vandergheynst P (2016) Convolutional neural networks on graphs with fast localized spectral filtering. In: Proceedings of the 29th international conference on neural information processing systems (NIPS-16), pp 3837–3845. arXiv:1606.09375v2
Kipf T, Welling M (2017) Semi-supervised classification with graph convolutional networks. In: Proceedings of the 5th international conference on learning representations (ICLR-17), arXiv:1609.02907
Hamilton W, Ying R, Leskovec J (2017) Inductive representation learning on large graphs. In: Proceedings of the 30th international conference on neural information processing systems (NIPS-17), pp 1024–1034. arXiv:1706.02216
Velickovic P, Cucurul G, Casanova A, Romero A, Lio P, Bengio Y (2017) Graph attention networks. In: Proceedings of the 5th international conference on learning representations (ICLR-17). arXiv:1710.10903
Yan S, Xiong Y, Lin D (2018) Spatial temporal graph convolutional networks for skeleton-based action recognition. In: Proceedings of the 32nd AAAI conference on artificial intelligence (AAAI-18), pp 7444–7452. arXiv:1801.07455
Yu B, Yin H, Zhu Z (2018) Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting. In: Proceedings of the twenty-seventh international joint conference on artificial intelligence (IJCAI-18), pp 3634–3640. arXiv:1709.04875v4
Wu Z, Pan S, Long G, Jiang J, Zhang C (2019) Graph wavenet for deep spatial-temporal graph modeling. In: Proceedings of the twenty-eighth international joint conference on artificial intelligence (IJCAI-19), pp 1907–1913. arXiv:1906.00121v1
Peng H, Wang H, Du B, Bhuiyan M, Ma H, Liu J, Wang L, Yang Z, Du L, Wang S, Yu S (2020) Spatial temporal incidence dynamic graph neural networks for traffic flow forecasting. Inf Sci 521:277–290. https://doi.org/10.1016/j.ins.2020.01.043
Zhang J, Chen F, Guo Y (2020) Multi-graph convolutional network for short-term passenger flow forecasting in urban rail transit. IET Intell Transp Syst 14(10):1210–1217. arXiv:2001.07512v3
Paparrizos J, Gravano L (2016) k-Shape: Efficient and accurate clustering of time series. Sigmoid Record 45(1):69–76. https://doi.org/10.1145/2949741.2949758
Lioutas V, Guo Y (2020) Time-aware large kernel convolution. In: Proceedings of the 37th international conference on machine learning (ICML-20), pp 6172–6183. arXiv:2002.03184
Zhang J, Zheng Y, Qi D (2017) Deep spatio-temporal residual networks for citywide crowd flows prediction. In: Proceedings of the 31nd AAAI conference on artificial intelligence (AAAI-17), pp 1655–1661. arXiv:1610.00081
Acknowledgements
This work is supported by the Natural Science Foundation of China (61866007), Natural Science Foundation of Guangxi Province (2018GXNSFDA138006), and the Guangxi Key Laboratory of Trusted Software (KX20202024).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Wei, Q., Qiu, Y. & Wen, Y. Cluster-based spatiotemporal dual self-adaptive network for short-term subway passenger flow forecasting. Appl Intell 52, 14137–14152 (2022). https://doi.org/10.1007/s10489-022-03305-w
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-022-03305-w