Research review on STGNN in traffic prediction: From model deconstruction to development path
Article Text (Baidu Translation)
-
摘要: 为厘清交通预测模型的发展路径，探索未来交通预测发展方向，通过系统化的文献分析方法确立了以时空图神经网络为主导的技术发展方向；基于时空图神经网络框架特征，构建了从数据预处理、动静态图构建、时空特征提取到特征融合的全流程分析体系；系统梳理了典型交通预测任务及其对应的开源数据集；归纳了基于拓扑关系、距离特性、相似度计算等静态图构建方法，以及动态图直接优化与特征优化等前沿图构建技术；从时间与空间两大维度总结分析了当前较新的时间特征建模与空间特征建模方法，并通过Graph WaveNet与DCRNN两大典型案例阐释了时空特征融合机制；针对深层网络训练中的梯度异常和性能衰退问题，总结了通用信息传递的解决方案；探索了对比学习、预训练机制、因果推理、混合专家模型等新兴技术与交通预测的结合路径。研究结果表明：无论是中国还是国外，时空图神经网络在交通预测的应用研究已经逐渐白热化，其中中国在统计中以1 671篇的发文量占据榜首；现有研究主要聚焦于提高模型对于时空特征的记忆能力与构建最优图结构，而这类优化方案已进入模型性能提升与效率提高的平衡期；在时间建模方面，现有研究仍在寻找计算效率与运行性能的平衡点，而空间建模则成为现有模型效率提升的主要阻碍；根据对过去研究的总结梳理，未来的突破方向或将集中于新型预测场景的开拓、模型可解释能力的提升、现实物理约束的添加、新型学习策略的创新以及工业级部署方案的探索，为智慧交通系统建设提供更坚实的技术支撑。Abstract: To clarify the development path of traffic prediction models and to explore future development directions of traffic prediction, a systematic literature analysis approach was adopted, and a technology development direction dominated by spatiotemporal graph neural networks (STGNN) was established. Based on the framework characteristics of STGNNs, a full-process analysis system was constructed, including data preprocessing, static and dynamic graph construction, spatiotemporal feature extraction, and feature fusion. Typical traffic prediction tasks and their corresponding open-source datasets were systematically reviewed. Static graph construction methods based on topological relationships, distance properties, and similarity calculations were summarized, and frontier graph construction technologies were summarized, including direct optimization of dynamic graphs and feature optimization. Current temporal feature modeling methods and spatial feature modeling methods were analyzed from the two dimensions of time and space, and the spatiotemporal feature fusion mechanism was illustrated through two typical cases of Graph WaveNet and DCRNN. For the problems of gradient anomalies and performance degradation in deep network training, general solutions based on information propagation were summarized. The integration paths of emerging technologies were explored, including contrastive learning, pre-training mechanisms, causal reasoning, and mixture-of-experts models with traffic prediction. Analysis results show that applications of STGNNs in traffic prediction have gradually intensified both domestically and internationally, and China ranks first with 1 671 publications in the statistics. Existing studies are mainly focused on improving model memory ability for spatio-temporal features and constructing optimal graph structures, and such optimization schemes have reached a balance between model performance and efficiency. In temporal modeling, a balance between computational efficiency and operational performance is still being explored, while spatial modeling has become the main obstacle to efficiency improvement of existing models. Based on the summary and review of previous studies, future breakthrough directions are expected to be concentrated on the exploration of novel prediction scenarios, improvement of model interpretability, incorporation of real-world physical constraints, innovation of learning strategies, and exploration of industrial-level deployment solutions, so that stronger technical support is provided for intelligent transportation systems.
-
图 6 交通节点之间的特殊空间关系
Figure 6. Special spatial relationships between transportation nodes
图 9 sin(x)、尖峰sin(x)、交替sin(x)时域图像及注意力分数
Figure 9. Sin(x), peak sin(x), alternating sin(x) time-domain images and attention scores
表 1 关键词共现频次排序
Table 1. Keyword co-occurrence frequency ranking
序号 CNKI WoS 关键词 频次 关键词 频次 1 深度学习 195 Deep learning 290 2 交通预测 100 Prediction 280 3 时空特征 84 Predictive models 258 4 智能交通 65 Traffic prediction 219 5 时空数据 54 Traffic flow prediction 187 6 时空特性 34 Neural network 184 7 图卷积 34 Graph convolutional network 150 8 时空预测 25 Attention mechanism 134 9 神经网络 23 Feature extraction 121 10 城市交通 23 Trajectory prediction 120 
下载: 导出CSV
表 2 主要研究国家分布
Table 2. Distribution of main research countries
国家 发文篇数 发文量排序 平均被引篇数 平均被引量排序 中国 1 671 1 8.01 7 美国 300 2 8.89 4 英格兰 101 3 8.38 6 澳大利亚 98 4 12.31 3 加拿大 73 5 12.51 2 印度 65 6 4.12 10 韩国 63 7 8.40 5 新加坡 59 8 27.59 1 日本 45 9 5.53 9 德国 38 10 7.44 8
下载: 导出CSV
表 3 不同交通预测任务可用公开数据集统计
Table 3. Statistics of different traffic prediction tasks using public datasets
预测任务 数据名称 数据介绍 部分应用模型 交通状态 PeMS PeMS-Bay PeMS-Bay为加州高速公路2017-01-01到2017-05-01交通速度数据集，包含325个节点，共52 116条数据，采样间隔为5 min DCRNN[27]
AutoSTS[47]
MTGNN[48]已处理数据： https://github.com/liyaguang/DCRNN (已征求作者同意)PeMS03 PeMS03为加州高速公路2018-09-01到2018-11-30交通数据，包括358个节点，共26 208条数据，常用数据为流量数据，采样间隔为5 min STGODE[49]
STSGCN[50]PeMS04 PeMS04为加州高速公路2018-01-01到2018-02-28交通数据，包括307个节点，共16 992条数据，常用数据为交通流量、交通速度以及占有率，采样间隔为5 min Z-GCNETs[51]
ASTGCN [52]
STFGCN[29]PeMS07 PeMS07为加州高速公路2017-05-01到2017-08-31交通数据，包括883个节点，共28 224条数据，常用数据为流量数据，采样间隔为5 min STGCN[53]
MVSTGCN[54]PeMS08 PeMS08为加州高速公路2016-07-01到2016-08-31交通数据，包括170个节点，共17 856条数据，常用数据为交通流量、交通速度以及占有率，采样间隔为5 min ST-AE[35]
StemGNN[55]数据源链接： http://pems.dot.ca.gov/ Metr-LA Metr-LA为洛杉矶地区2012-03-01到2012-06-30交通速度数据集，包含207个节点，共24 272条数据，采样间隔为5 min GWN [56]
LightCTS[57]数据源链接： https://www.metro.net/
已处理数据：https://github.com/liyaguang/DCRNN (已征求作者同意)EXPY-TKY EXPY-TKY为东京地区2021-12-01到2021-12-31交通速度数据，包含1 843条链接，共13 248条数据，采样间隔为10 min MegaCRN[58]
TESTAM[59]数据源链接: https://en.wikipedia.org/wiki/Shuto_Expressway
已处理数据链接：https://github.com/deepkashiwa20/MegaCRN (MIT License)交通需求行人流量 TaxiNYC TaxiNYC为纽约出租车2011-01-01到2016-06-30期间3 500万条行程数据，每条行程数据包括上车时间、下车时间、上车经纬度、下车经纬度、行程距离 CCRNN[42]
MDSR[60]数据源链接： https://www1.nyc.gov/site/tlc/about/tle-trip-record-data.page TaxiBJ TaxiBJ包含北京出租车2013-07-01到2017-10-30、2014-03-01到2014-06-30、2015-03-01到2015-06-30、2015-11-01到2016-6-30四个时间段超过34 000辆出租车行程数据 ST-ResNet[61] 处理数据链接： https://github.com/topazape/ST-ResNet (MIT License)BikeNYC BikeNYC包含纽约超过6 800辆自行车在2014-04-01到2014-09-30期间的行程数据，包括出行时长、起止站点ID、起止时间 ST-ResNet[61]
STMN[62]数据源链接： https://www.citibikenyc.com/system-data
处理数据链接：https://github.com/topazape/ST-ResNet (MIT License)交通事故 NYC Accident NYC Accident为多个时间段交通事故数据，如GSNet使用为2013-01-01到2013-12-31数据，RiskSeq使用为2017-01-01到2017-05-31数据，包括交通事故记录，出租车订单、兴趣点以及天气等多类数据 GSNet[39]
Riskoracle[63]数据源链接： https://opendata.cityofnewyork.us/
已处理数据：https://github.com/Echohhhhhh/GSNet (已征求作者同意)Chicago Accident Chicago Accident为2016-02-01到2016-09-30交通事故数据，包含44 000+交通事故记录以及1 744 000+出租车订单等多类数据 GSNet[39] 数据源链接： https://data.cityofchicago.org
已处理数据：https://github.com/Echohhhhhh/GSNet (已征求作者同意)车辆轨迹行程时间 Porto dataset 包含葡萄牙波尔图2013-01-07到2014-06-30的442辆出租车超过300 000条轨迹数据，其内容包括司机标号、行程开始时间、行程总时间等9种类型数据 STGNN-TTE[64]
START[65]数据源链接： https://www.kaggle.com/datasets/crailtap/taxi-trajectory T-Drive T-Drive [66-67]包含北京地区2008-02-02到2008-02-08期间 10 357辆出租车约1 500条GPS轨迹数据 PreCLN[68] 链接： https://www.microsoft.com/en-us/research/publication/t-drive-trajectory-data-sample/ 车辆轨迹 NGSIM NGSIM包含I-80和US-1012个高速公路段以及兰克希姆大道和桃树街2个干道段9 206条车辆轨迹数据，内容含车辆行驶状态、车辆坐标、车辆长度等多种数据，采样频率为10 Hz CDSTraj[69]
STDAN[70]数据源链接： https://data.transportation.gov/Automobiles/Next-Generation-Simulation-NGSIM-Vehicle-Trajector/8ect-6jqj/about_data HighD[71] HighD为德国科隆6条约420 m高速公路通过无人机采集的包含 110 000+辆汽车，总里程45 000 km的行驶数据，其中有约5 600条车辆变道记录，采样频率为25 Hz CDSTraj[69]
STDAN[70]数据源链接： https://levelxdata.com/highd-dataset/ Argoverse[72] Argoverse包括3D追踪与运动轨迹2部分，其中3D追踪包含LiDAR和视觉传感器从113个场景采集的数据，轨迹数据包含1 000+小时驾驶时间提取的324 557条轨迹数据 LaneGCN[73]
STGACN[74]数据源链接： https://www.argoverse.org/ https://outreach.didichuxing.com/en 可对DiDi Chengdu与DiDi Xi'an出租车GPS轨迹数据进行申请[75]
下载: 导出CSV
表 4 近年最先进模型预测效果统计
Table 4. Recent statistics on the prediction performance of state-of-the-art models
对比领域 DCRNN (2018年) ASTGCN (2019年) Graph Wave Net(2019年) GMAN (2020年) GTS (2021年) DDSTGCN (2022年) STAEformer (2023年) HimNet (2024年) 时间建模 GRU Former TCN Former CNN TCN Former GRU 空间建模 空域 GCN 谱域 GCN 空域 GCN GAT 空域 GCN 空域 GCN 无 空域 GCN 融合策略 耦合式 堆叠式 堆叠式 堆叠式 堆叠式 堆叠式 堆叠式 耦合式 Metr-LA RMSE 7.59 7.75 7.37 7.35 7.44 7.13 7.02 7.22 MAE 3.60 3.83 3.53 3.44 3.59 3.44 3.34 3.37 MAPE% 10.5 10.8 10.0 10.1 10.3 9.7 9.7 9.8 PeMS-Bay RMSE 4.74 4.81 4.52 1.92 4.60 4.37 4.34 4.32 MAE 2.07 2.15 1.95 4.49 2.06 1.89 1.88 1.84 MAPE/% 4.9 5.1 4.6 4.5 4.9 4.5 4.4 4.3 PeMS04 RMSE 31.26 35.22 29.92 31.60 32.95 31.05 30.18 29.88 MAE 19.63 22.93 18.53 19.14 20.96 19.61 18.22 18.14 MAPE/% 13.6 16.6 12.9 13.2 14.7 13.7 12.0 12.0 PeMS08 RMSE 24.17 28.16 23.39 24.92 26.08 24.16 23.25 23.22 MAE 15.22 18.61 14.40 15.31 16.49 15.30 13.46 13.57 MAPE/% 10.2 13.1 9.2 10.1 10.5 10.5 8.9 9.0
下载: 导出CSV
-
[1] 冯晓, 陈思龙. 改善城市道路机动车排放污染的智能交通手段[J]. 交通运输工程学报, 2002, 2(2): 73-77. doi: 10.3321/j.issn:1671-1637.2002.02.018FENG Xiao, CHEN Si-long. An ITS method to decrease motor-vehicle pollution in urban area[J]. Journal of Traffic and Transportation Engineering, 2002, 2(2): 73-77. doi: 10.3321/j.issn:1671-1637.2002.02.018 [2] FENG K R, LIN N. Reconstructing and analyzing the traffic flow during evacuation in Hurricane Irma (2017)[J]. Transportation Research Part D: Transport and Environ-ment, 2021, 94: 102788. doi: 10.1016/j.trd.2021.102788 [3] 臧华, 彭国雄. 高速道路异常状况下车辆排队长度的预测模型[J]. 交通与计算机, 2003, 21(3): 10-12.ZANG Hua, PENG Guo-xiong. A forecast model of queuing length in expressway emergency[J]. Computer and Commu-nications, 2003, 21(3): 10-12. [4] LARTEY J D. Predicting traffic congestion: A queuing perspective[J]. Open Journal of Modelling and Simulation, 2014, 2(2): 57-66. doi: 10.4236/ojmsi.2014.22008 [5] 王永泉, 陈花玲, 毛文雄. 基于跟驰理论的道路交通噪声预测模型研究[J]. 系统仿真学报, 2004, 16(11): 2413-2416.WANG Yong-quan, CHEN Hua-ling, MAO Wen-xiong. Research on road traffic noise prediction model based on car-following theory[J]. Acta Simulata Systematica Sinica, 2004, 16(11): 2413-2416. [6] DAGANZO C F. The cell transmission model: A dynamic representation of highway traffic consistent with the hydrodynamic theory[J]. Transportation Research Part B: Methodological, 1994, 28(4): 269-287. doi: 10.1016/0191-2615(94)90002-7 [7] XU X, ZHANG L L, KONG Q, et al. Enhanced-historical average for long-term prediction[C]//IEEE. 2022 2nd International Conference on Computer, Control and Robotics (ICCCR). New York: IEEE, 2022: 115-119. [8] HAMED M M, AL-MASAEID H R, SAID Z M B. Short-term prediction of traffic volume in urban arterials[J]. Journal of Transportation Engineering, 1995, 121(3): 249-254. doi: 10.1061/(ASCE)0733-947X(1995)121:3(249) [9] SHAHRIARI S, GHASRI M, SISSON S A, et al. Ensemble of ARIMA: Combining parametric and bootstrapping technique for traffic flow prediction[J]. Transportmetrica A: Transport Science, 2020, 16(3): 1552-1573. doi: 10.1080/23249935.2020.1764662 [10] CORTES C, VAPNIK V. Support-vector networks[J]. Machine Learning, 1995, 20(3): 273-297. doi: 10.1023/A:1022627411411 [11] BREIMAN L. Random forests[J]. Machine Learning, 2001, 45(1): 5-32. doi: 10.1023/A:1010933404324 [12] COVER T, HART P. Nearest neighbor pattern classification[J]. IEEE Transactions on Information Theory, 1967, 13(1): 21-27. doi: 10.1109/TIT.1967.1053964 [13] WU C H, HO J M, LEE D T. Travel-time prediction with support vector regression[J]. IEEE Transactions on Intelligent Transportation Systems, 2004, 5(4): 276-281. doi: 10.1109/TITS.2004.837813 [14] RUMELHART D E, HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors[J]. Nature, 1986, 323: 533-536. doi: 10.1038/323533a0 [15] BAO X X, JIANG D, YANG X F, et al. An improved deep belief network for traffic prediction considering weather factors[J]. Alexandria Engineering Journal, 2021, 60(1): 413-420. doi: 10.1016/j.aej.2020.09.003 [16] ZHANG W B, YU Y H, QI Y, et al. Short-term traffic flow prediction based on spatio-temporal analysis and CNN deep learning[J]. Transportmetrica A: Transport Science, 2019, 15(2): 1688-1711. doi: 10.1080/23249935.2019.1637966 [17] WILLIAMS R J, ZIPSER D. A learning algorithm for continually running fully recurrent neural networks[J]. Neural Computation, 1989, 1(2): 270-280. doi: 10.1162/neco.1989.1.2.270 [18] HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780. doi: 10.1162/neco.1997.9.8.1735 [19] CHO K, VAN MERRIENBOER B, GULCEHRE C, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation[C]//ACL. Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP). Doha: ACL, 2014: 1724-1734. [20] ASHISH V, NOAM S, NIKI P, et al. Attention is all you need[C]//NeurIPS Foundation. Conference on Neural Infor-mation Processing Systems. Cambridge: MIT Press, 2017: 6000-6010. [21] 马飞, 杨治杰, 王江博, 等. 基于气象-交通多通道数据融合的短时交通流速度预测模型[J]. 交通运输工程学报, 2024, 24(6): 183-196. doi: 10.19818/j.cnki.1671-1637.2024.06.013MA Fei, YANG Zhi-jie, WANG Jiang-bo, et al. Short-term traffic flow speed prediction model based on meteorological-traffic multi-channel data fusion[J]. Journal of Traffic and Transportation Engineering, 2024, 24(6): 183-196. doi: 10.19818/j.cnki.1671-1637.2024.06.013 [22] 胡立伟, 侯智, 赵雪亭, 等. 基于交通事故文本挖掘的高速公路行车风险预测模型改进研究[J/OL]. 西南交通大学学报, 2025, http://kns.cnki.net/kcms/detail/51.1277.U.20240509.1523.010.html .HU Li-wei, HOU Zhi, ZHAO Xue-ting, et al. Research on improvement of highway traffic risk prediction model based on traffic accident text mining[J/OL]. Journal of Southwest Jiaotong University, 2025,http://kns.cnki.net/kcms/detail/51.1277.U.20240509.1523.010.html .[23] MA C X, ZHAO Y P, DAI G W, et al. A novel STFSA-CNN-GRU hybrid model for short-term traffic speed prediction[J]. IEEE Transactions on Intelligent Trans-portation Systems, 2023, 24(4): 3728-3737. doi: 10.1109/TITS.2021.3117835 [24] BEECHE C, SINGH J P, LEADER J K, et al. Super U-Net: A modularized generalizable architecture[J]. Pattern Recog-nition, 2022, 128: 108669. doi: 10.1016/j.patcog.2022.108669 [25] ZHANG Z B, WU S, JIANG D W, et al. BERT-JAM: Maximizing the utilization of BERT for neural machine translation[J]. Neurocomputing, 2021, 460: 84-94. doi: 10.1016/j.neucom.2021.07.002 [26] 杨国亮, 习浩, 龚家仁, 等. 基于Transformer的短时交通流时空预测[J]. 计算机应用与软件, 2024, 41(3): 169-173, 225.YANG Guo-liang, XI Hao, GONG Jia-ren, et al. Short term traffic flow forecasting based on transformer[J]. Computer Applications and Software, 2024, 41(3): 169-173, 225. [27] LI Y G, YU R, CYRUS S, et al. Diffusion convolutional recurrent neural network: Data-driven traffic forecasting[C]//YOSHUA B, YANN L. International Conference on Learning Representations. Portland: OpenReview. net, 2018: 1-16. [28] DONG Z, JIANG R H, GAO H T, et al. Heterogeneity-informed meta-parameter learning for spatiotemporal time series forecasting[C]//ACM. Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York: ACM, 2024: 631-641. [29] LI H, LIU J, HAN S Y, et al. STFGCN: Spatial-temporal fusion graph convolutional network for traffic prediction[J]. Expert Systems with Applications, 2024, 255: 124648. doi: 10.1016/j.eswa.2024.124648 [30] WANG S Z, CAO J N, YU P S. Deep learning for spatio-temporal data mining: A survey[J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 34(8): 3681-3700. doi: 10.1109/TKDE.2020.3025580 [31] 崔建勋, 要甲, 赵泊媛. 基于深度学习的短期交通流预测方法综述[J]. 交通运输工程学报, 2024, 24(2): 50-64. doi: 10.19818/j.cnki.1671-1637.2024.02.003CUI Jian-xun, YAO Jia, ZHAO Bo-yuan. Review on short-term traffic flow prediction methods based on deep learning[J]. Journal of Traffic and Transportation Engineering, 2024, 24(2): 50-64. doi: 10.19818/j.cnki.1671-1637.2024.02.003 [32] 邹慧琪, 史彬泽, 宋凌云, 等. 基于图神经网络的复杂时空数据挖掘方法综述[J]. 软件学报, 2025, 36(4): 1811-1843.ZOU Hui-qi, SHI Bin-ze, SONG Ling-yun, et al. Survey on complex spatio-temporal data mining methods based on graph neural networks[J]. Journal of Software, 2025, 36(4): 1811-1843. [33] 户佐安, 邓锦程, 韩金丽, 等. 图神经网络在交通预测中的应用综述[J]. 交通运输工程学报, 2023, 23(5): 39-61. doi: 10.19818/j.cnki.1671-1637.2023.05.003HU Zuo-an, DENG Jin-cheng, HAN Jin-li, et al. Review on application of graph neural network in traffic prediction[J]. Journal of Traffic and Transportation Engineering, 2023, 23(5): 39-61. doi: 10.19818/j.cnki.1671-1637.2023.05.003 [34] 熊章友, 李卫军, 朱晓娟, 等. 基于深度学习的短时交通流预测研究综述[J]. 计算机工程与应用, 2025, 61(11): 67-82.XIONG Zhang-you, LI Wei-jun, ZHU Xiao-juan, et al. Short-term traffic flow prediction based on deep learning[J]. Computer Engineering and Applications, 2025, 61(11): 67-82. [35] LIU M Z, ZHU T Y, YE J C, et al. Spatio-temporal autoencoder for traffic flow prediction[J]. IEEE Transac-tions on Intelligent Transportation Systems, 2023, 24(5): 5516-5526. doi: 10.1109/TITS.2023.3243913 [36] ABDELRAOUF A, ABDEL-ATY M, YUAN J H. Utilizing attention-based multi-encoder-decoder neural networks for freeway traffic speed prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(8): 11960-11969. doi: 10.1109/TITS.2021.3108939 [37] WU X L, ZHANG D L, GUO C J, et al. AutoCTS: Automated correlated time series forecasting[J]. Proceedings of the VLDB Endowment, 2022, 15(4): 971-983. [38] ZOU G J, LAI Z L, MA C X, et al. When will we arrive a novel multi-task spatio-temporal attention network based on individual preference for estimating travel time[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24(10): 11438-11452. doi: 10.1109/TITS.2023.3276916 [39] WANG B B, LIN Y F, GUO S N, et al. GSNet: Learning spatial-temporal correlations from geographical and semantic aspects for traffic accident risk forecasting[C]//AAAI. Proceedings of the AAAI Conference on Artificial Intel-ligence. Washington DC: AAAI Press, 2021: 4402-4409. [40] WANG Z N, JIANG R H, XUE H, et al. Event-aware multimodal mobility nowcasting[C]//AAAI. Proceedings of the AAAI Conference on Artificial Intelligence. Washington DC: AAAI Press, 2022: 4228-4236. [41] 贾兴利, 李双庆, 杨宏志, 等. 基于ATT-LSTM模型的高速公路交通事件持续时长预测[J]. 交通信息与安全, 2022, 40(5): 61-69.JIA Xing-li, LI Shuang-qing, YANG Hong-zhi, et al. Prediction of the duration of freeway traffic incidents based on an ATT-LSTM model[J]. Journal of Transport Information and Safe, 2022, 40(5): 61-69. [42] YE J C, SUN L L, DU B W, et al. Coupled layer-wise graph convolution for transportation demand prediction[C]//AAAI. Proceedings of the AAAI Conference on Artificial Intelligence. Washington DC: AAAI Press, 2021: 4616-4625. [43] WANG S Y, ZHUGE C X, SHAO C F, et al. Short-term electric vehicle charging demand prediction: A deep learning approach[J]. Applied Energy, 2023, 340: 121032. doi: 10.1016/j.apenergy.2023.121032 [44] LIANG Y B, DING F Y, HUANG G, et al. Deep trip generation with graph neural networks for bike sharing system expansion[J]. Transportation Research Part C: Emerging Technologies, 2023, 154: 104241. doi: 10.1016/j.trc.2023.104241 [45] HUANG Y J, DU J T, YANG Z R, et al. A survey on trajectory-prediction methods for autonomous driving[J]. IEEE Transactions on Intelligent Vehicles, 2022, 7(3): 652-674. doi: 10.1109/TIV.2022.3167103 [46] CHEN G X, HU L, ZHANG Q S, et al. ST-LSTM: Spatio-temporal graph based long short-term memory network for vehicle trajectory prediction[C]//IEEE. 2020 IEEE Inter-national Conference on Image Processing (ICIP). New York: IEEE, 2020: 608-612. [47] LI F X, YAN H, JIN G Y, et al. Automated spatio-temporal synchronous modeling with multiple graphs for traffic prediction[C]//ACM. Proceedings of the 31st ACM Inter-national Conference on Information & Knowledge Manage-ment. New York: ACM, 2022: 1084-1093. [48] WU Z H, PAN S R, LONG G D, et al. Connecting the dots: Multivariate time series forecasting with graph neural networks[C]//ACM. Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2020: 753-763. [49] FANG Z, LONG Q Q, SONG G J, et al. Spatial-temporal graph ODE networks for traffic flow forecasting[C]//ACM. Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. New York: ACM, 2021: 364-373. [50] SONG C, LIN Y F, GUO S N, et al. Spatial-temporal synchronous graph convolutional networks: A new frame-work for spatial-temporal network data forecasting[C]//AAAI. Proceedings of the AAAI Conference on Artificial Intelligence. Washington DC: AAAI Press, 2020: 914-921. [51] CHEN Y Z, DOMINGUEZ I S, GEL Y R. Z-GCNETs: Time zigzags at graph convolutional networks for time series forecasting[C]//ACM. International Conference on Machine Learning. New York: ACM, 2021: 1684-1694. [52] GUO S N, LIN Y F, FENG N, et al. Attention based spatial-temporal graph convolutional networks for traffic flow forecasting[C]//AAAI. Proceedings of the AAAI Confe-rence on Artificial Intelligence. Washington DC: AAAI Press, 2019: 922-929. [53] YU B, YIN H T, ZHU Z X. Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting[C]//AAAI. Proceedings of the 27th International Joint Conference on Artificial Intelligence. Washington DC: AAAI Press, 2018: 3634-3640. [54] 赵文竹, 袁冠, 张艳梅, 等. 多视角融合的时空动态GCN城市交通流量预测[J]. 软件学报, 2024, 35(4): 1751-1773.ZHAO Wen-zhu, YUAN Guan, ZHANG Yan-mei, et al. Multi-view fused spatial-temporal dynamic GCN for urban traffic flow prediction[J]. Journal of Software, 2024, 35(4): 1751-1773. [55] CAO D F, WANG Y J, DUAN J Y, et al. Spectral temporal graph neural network for multivariate time-series forecasting[C]//NeurIPS Foundation. Conference on Neural Infor-mation Processing Systems. Cambridge: MIT Press, 2020: 17766-17778. [56] WU Z H, PAN S R, LONG G D, et al. Graph WaveNet for deep spatial-temporal graph modeling[C]//AAAI. Proceedings of the 28th International Joint Conference on Artificial Intelligence. Washington DC: AAAI Press, 2019: 1907-1913. [57] LAI Z C, ZHANG D L, LI H, et al. LightCTS*: Light-weight correlated time series forecasting enhanced with model distillation[J]. IEEE Transactions on Knowledge and Data Engineering, 2024, 36(12): 8695-8710. doi: 10.1109/TKDE.2024.3424451 [58] JIANG R H, WANG Z N, YONG J W, et al. Spatio-temporal meta-graph learning for traffic forecasting[C]//AAAI. Proceedings of the AAAI Conference on Artificial Intelligence. Washington DC: AAAI Press, 2023: 8078-8086. [59] LEE H, KO S. TESTAM: A time-enhanced spatio-temporal attention model with mixture of experts[C]//YOSHUA B, YANN L. International Conference on Learning Represen-tations. Portland: OpenReview. net, 2024: 1-19. [60] LIU D C, WANG J, SHANG S, et al. MSDR: Multi-step dependency relation networks for spatial temporal forecasting[C]//ACM. Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York: ACM, 2022: 1042-1050. [61] ZHANG J B, ZHENG Y, QI D K. Deep spatio-temporal residual networks for citywide crowd flows prediction[C]//AAAI. AAAI Conference on Artificial Intelligence. Washington DC: AAAI Press, 2017: 1-7. [62] LI X Y, XU Y, CHEN Q, et al. Short-term forecast of bicycle usage in bike sharing systems: A spatial-temporal memory network[J]. IEEE Transactions on Intelligent Trans-portation Systems, 2022, 23(8): 10923-10934. doi: 10.1109/TITS.2021.3097240 [63] ZHOU Z Y, WANG Y, XIE X K, et al. RiskOracle: A minute-level citywide traffic accident forecasting framework[C]//AAAI. Proceedings of the AAAI Conference on Artificial Intelligence. Washington DC: AAAI Press, 2020: 1258-1265. [64] JIN G Y, WANG M, ZHANG J L, et al. STGNN-TTE: Travel time estimation via spatial-temporal graph neural network[J]. Future Generation Computer Systems, 2022, 126: 70-81. doi: 10.1016/j.future.2021.07.012 [65] JIANG J W, PAN D Y, REN H X, et al. Self-supervised trajectory representation learning with temporal regularities and travel semantics[C]//IEEE. 2023 IEEE 39th Inter-national Conference on Data Engineering (ICDE). New York: IEEE, 2023: 843-855. [66] YUAN J, ZHENG Y, XIE X, et al. Driving with knowledge from the physical world[C]//ACM. Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2011: 316-324. [67] YUAN J, ZHENG Y, ZHANG C Y, et al. T-drive: Driving directions based on taxi trajectories[C]//ACM. Proceedings of the 18th SIGSPATIAL International Conference on Advances in Geographic Information Systems. New York: ACM, 2010: 99-108. [68] YAN B Q, ZHAO G, SONG L X, et al. PreCLN: Pretrained-based contrastive learning network for vehicle trajectory prediction[J]. World Wide Web, 2023, 26(4): 1853-1875. doi: 10.1007/s11280-022-01121-3 [69] LIAO H C, LI X L, LI Y K, et al. Characterized diffusion and spatial-temporal interaction network for trajectory prediction in autonomous driving[C]//AAAI. International Joint Conference on Artificial Intelligence. Washington DC: AAAI Press, 2024: 1-9. [70] CHEN X B, ZHANG H J, ZHAO F, et al. Intention-aware vehicle trajectory prediction based on spatial-temporal dynamic attention network for Internet of vehicles[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(10): 19471-19483. doi: 10.1109/TITS.2022.3170551 [71] KRAJEWSKI R, BOCK J, KLOEKER L, et al. The highD dataset: A drone dataset of naturalistic vehicle trajectories on German highways for validation of highly automated driving systems[C]//IEEE. 2018 21st International Conference on Intelligent Transportation Systems (ITSC). New York: IEEE, 2018: 2118-2125. [72] CHANG M-F, LAMBERT J, SANGKLOY P, et al. Argoverse: 3D tracking and forecasting with rich maps[C]//IEEE. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New York: IEEE, 2019: 8740-8749. [73] LIANG M, YANG B, HU R, et al. Learning lane graph representations for motion forecasting[C]//Springer. Proceedings of 16th European Conference on Computer Vision (ECCV). Munich: Springer, 2020: 541-556. [74] 袁静, 夏英. 基于时空图注意力卷积神经网络的车辆轨迹预测[J]. 计算机科学, 2024, 51(12): 157-165.YUAN Jing, XIA Ying. Vehicle trajectory prediction based on spatial-temporal graph attention convolutional network[J]. Computer Science, 2024, 51(12): 157-165. [75] CHANG Y C, QI J Z, LIANG Y X, et al. Contrastive trajectory similarity learning with dual-feature attention[C]//IEEE. 2023 IEEE 39th International Conference on Data Engineering (ICDE). New York: IEEE, 2023: 2933-2945. [76] 付相君. 基于路网拓扑结构的相关路段流量预测研究[D]. 重庆: 重庆交通大学, 2020.FU Xiang-jun. Research on traffic forecasting of related sections based on topological structure of road network[D]. Chongqing: Chongqing Jiaotong University, 2020. [77] 温振国. 基于复杂网络理论的陕西省高速公路网结构特性及鲁棒性研究[D]. 西安: 长安大学, 2019.WEN Zhen-guo. Study on structural characteristics and robustness of Shaanxi expressway network based on complex network theory[D]. Xi'an: Chang'an University, 2019. [78] 陈希, 钱江海, 韩定定. L空间和P空间模型下的树状结构网络[J]. 计算机应用研究, 2015, 32(1): 45-47.CHEN Xi, QIAN Jiang-hai, HAN Ding-ding. Tree network under space L and space P model[J]. Application Research of Computers, 2015, 32 (1): 45-47. [79] WANG X Y, MA Y, WANG Y Q, et al. Traffic flow prediction via spatial temporal graph neural network[C]//ACM. Proceedings of The Web Conference 2020. New York: ACM, 2020: 1082-1092. [80] GAO A, ZHENG L J, WANG Z X, et al. Attention based short-term metro passenger flow prediction[M]. Munich: Springer International Publishing, 2021. [81] GUO K, HU Y L, QIAN Z, et al. Optimized graph convolution recurrent neural network for traffic prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(2): 1138-1149. doi: 10.1109/TITS.2019.2963722 [82] 孙俊, 潘玉君, 和瑞芳, 等. 地理学第一定律之争及其对地理学理论建设的启示[J]. 地理研究, 2012, 31(10): 1749-1763.SUN Jun, PAN Yu-jun, HE Rui-fang, et al. The enligh-tenment of geographical theories construction from the first law of geography and its debates[J]. Geographical Research, 2012, 31(10): 1749-1763. [83] 谭会生, 杨威, 严舒琪. 交通速度预测时空图卷积网络及其FPGA实现研究[J]. 电子测量技术, 2024, 47(18): 108-119.TAN Hui-sheng, YANG Wei, YAN Shu-qi. Research on spatio-temporal graph convolutional network for traffic speed prediction and their FPGA implementation[J]. Electronic Measurement Technology, 2024, 47(18): 108-119. [84] ZHAO L, SONG Y J, ZHANG C, et al. T-GCN: A temporal graph convolutional network for traffic prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(9): 3848-3858. doi: 10.1109/TITS.2019.2935152 [85] 祁朵, 毛政元. 基于自适应时序剖分与KNN的短时交通流量预测[J]. 地球信息科学学报, 2022, 24(2): 339-351.QI Duo, MAO Zheng-yuan. Short-term traffic flow predic-tion based on adaptive time slice and KNN[J]. Journal of Geo-information Science, 2022, 24(2): 339-351. [86] LV M Q, HONG Z X, CHEN L, et al. Temporal multi-graph convolutional network for traffic flow prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(6): 3337-3348. doi: 10.1109/TITS.2020.2983763 [87] LI P, WANG S, ZHAO H T, et al. IG-net: An interaction graph network model for metro passenger flow forecasting[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24(4): 4147-4157. doi: 10.1109/TITS.2023.3235805 [88] XU G, LI Y G, WANG L Y, et al. Spatiotemporal multi-graph convolution network for ride-hailing demand forecasting[C]//AAAI. Proceedings of the AAAI Conference on Artificial Intelligence. Washington DC: AAAI Press, 2019: 3656-3663. [89] LIU L B, CHEN J W, WU H F, et al. Physical-virtual collaboration modeling for intra- and inter-station metro ridership prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(4): 3377-3391. doi: 10.1109/TITS.2020.3036057 [90] 包银鑫, 曹阳, 施佺. 基于改进时空残差卷积神经网络的城市路网短时交通流预测[J]. 计算机应用, 2022, 42(1): 258-264.BAO Yin-xin, CAO Yang, SHI Quan. Improved spatio-temporal residual convolutional neural network for urban road network short-term traffic flow prediction[J]. Journal of Computer Applications, 2022, 42(1): 258-264. [91] CHAI D, WANG L Y, YANG Q. Bike flow prediction with multi-graph convolutional networks[C]//ACM. Proceedings of the 26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. New York: ACM, 2018: 397-400. [92] 濮稳稳. 多类别交通参与者轨迹预测算法研究[D]. 长沙: 湖南大学, 2021.PU Wen-wen. Research on trajectory prediction algorithm of multi-class traffic participants[D]. Changsha: Hunan University, 2021. [93] SHAO W, JIN Z L, WANG S, et al. Long-term spatio-temporal forecasting via dynamic multiple-graph attention[C]//AAAI. Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence. Washington DC: AAAI Press, 2022: 1-8. [94] LIU R W, LIANG M H, NIE J T, et al. STMGCN: Mobile edge computing-empowered vessel trajectory prediction using spatio-temporal multigraph convolutional network[J]. IEEE Transactions on Industrial Informatics, 2022, 18(11): 7977-7987. doi: 10.1109/TII.2022.3165886 [95] LUO X L, ZHU C J, ZHANG D T, et al. Dynamic graph convolutional network with attention fusion for traffic flow prediction[C]//Springer. Proceedings of 26th European Conference on Artificial Intelligence (ECAI). Munich: Springer, 2023: 1-8. [96] MA Y, LOU H J, YAN M, et al. Spatio-temporal fusion graph convolutional network for traffic flow forecasting[J]. Information Fusion, 2024, 104: 102196. doi: 10.1016/j.inffus.2023.102196 [97] ZHANG W, ZHU F H, LV Y S, et al. AdapGL: An adaptive graph learning algorithm for traffic prediction based on spatiotemporal neural networks[J]. Transportation Research Part C: Emerging Technologies, 2022, 139: 103659. doi: 10.1016/j.trc.2022.103659 [98] QU Y H, JIA X L, GUO J H, et al. MSSTGNN: Multi-scaled spatio-temporal graph neural networks for short- and long-term traffic prediction[J]. Knowledge-based Systems, 2024, 306: 112716. doi: 10.1016/j.knosys.2024.112716 [99] CHAUHAN V K, ZHOU J D, LU P, et al. A brief review of hypernetworks in deep learning[J]. Artificial Intelligence Review, 2024, 57(9): 250. doi: 10.1007/s10462-024-10862-8 [100] PENG Y F, GUO Y Y, HAO R, et al. Network traffic prediction with attention-based spatial-temporal graph network[J]. Computer Networks, 2024, 243: 110296. doi: 10.1016/j.comnet.2024.110296 [101] SHIN Y, YOON Y. PGCN: Progressive graph convolutional networks for spatial-temporal traffic forecasting[J]. IEEE Transactions on Intelligent Transportation Systems, 2024, 25(7): 7633-7644. doi: 10.1109/TITS.2024.3349565 [102] JIN G Y, LIU C X, XI Z X, et al. Adaptive dual-view WaveNet for urban spatial-temporal event prediction[J]. Information Sciences, 2022, 588: 315-330. doi: 10.1016/j.ins.2021.12.085 [103] 吴永庆, 姜正宇. 基于解耦合动态时空卷积循环网络的交通流预测[J/OL]. 计算机工程, 2025, DOI: 10.19678/j.issn.1000-3428.0070319 .WU Yong-qing, JIANG Zheng-yu. Traffic flow prediction based on decoupled dynamic spatial-temporal convolutional recurrent network [J/OL]. Computer Engineering, 2025, DOI:10.19678/j.issn.1000-3428.0070319 .[104] TA X X, LIU Z H, HU X, et al. Adaptive spatio-temporal graph neural network for traffic forecasting[J]. Knowledge-Based Systems, 2022, 242: 108199. doi: 10.1016/j.knosys.2022.108199 [105] LV Z Q, WANG X T, CHENG Z S, et al. ST-TDCN: A two-channel tree-structure spatial-temporal convolutional network model for traffic velocity prediction[J]. Expert Systems with Applications, 2024, 257: 125053. doi: 10.1016/j.eswa.2024.125053 [106] WANG F, SUN J M. Survey on distance metric learning and dimensionality reduction in data mining[J]. Data Mining and Knowledge Discovery, 2015, 29(2): 534-564. doi: 10.1007/s10618-014-0356-z [107] WOJKE N, BEWLEY A. Deep cosine metric learning for person re-identification[C]//IEEE. 2018 IEEE Winter Conference on Applications of Computer Vision (WACV). New York: IEEE, 2018: 748-756. [108] CHEN Y, WU L F, ZAKI M J. Iterative deep graph learning for graph neural networks: Better and robust node embeddings[C]//NeurIPS Foundation. Proceedings of 34th Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2020: 19314-19326. [109] LI R Y, WANG S, ZHU F Y, et al. Adaptive graph convolutional neural networks[C]//AAAI. Proceedings of the AAAI Conference on Artificial Intelligence. Washington DC: AAAI Press, 2018: 3546-3553. [110] LIU H Y, YANG D H, LIU X Z, et al. TodyNet: Temporal dynamic graph neural network for multivariate time series classification[J]. Information Sciences, 2024, 677: 120914. doi: 10.1016/j.ins.2024.120914 [111] LUCA F, MATHIAS N, MASSIMILIANO P, et al. Learning discrete structures for graph neural networks[C]//ACM. Proceedings of 36th International Conference on Machine Learning. New York: ACM, 2019: 1972-1982. [112] ZHENG C, ZONG B, CHENG W, et al. Robust graph representation learning via neural sparsification[C]//ACM. Proceedings of 37th International Conference on Machine Learning. New York: ACM, 2020: 11458-11468. [113] JIN G Y, LIANG Y X, FANG Y C, et al. Spatio-temporal graph neural networks for predictive learning in urban computing: A survey[J]. IEEE Transactions on Knowledge and Data Engineering, 2024, 36(10): 5388-5408. doi: 10.1109/TKDE.2023.3333824 [114] GUO K, HU Y L, SUN Y F, et al. Hierarchical graph convolution network for traffic forecasting[C]//AAAI. Proceedings of the AAAI Conference on Artificial Intel-ligence. Washington DC: AAAI Press, 2021: 151-159. [115] AGARWAL S, SAWHNEY R, THAKKAR M, et al. THINK: Temporal hypergraph hyperbolic network[C]//IEEE. 2022 IEEE International Conference on Data Mining (ICDM). New York: IEEE, 2022: 849-854. [116] 侯越, 张鑫, 袭著涛, 等. 特殊路网拓扑解构下的时空异质化交通流预测[J/OL]. 铁道科学与工程学报, 2025, 22(7): 2932-2945.HOU Yue, ZHANG Xin, XI Zhu-tao, et al. Spatio-temporal heterogeneous traffic flow prediction under special road network topology deconstruction[J/OL]. Journal of Railway Science and Engineering, 2025, 22(7): 2932-2945. [117] WANG J C, ZHANG Y, WEI Y, et al. Metro passenger flow prediction via dynamic hypergraph convolution networks[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(12): 7891-7903. doi: 10.1109/TITS.2021.3072743 [118] ZHAO Y S, LUO X, JU W, et al. Dynamic hypergraph structure learning for traffic flow forecasting[C]//IEEE. 2023 IEEE 39th International Conference on Data Engi-neering (ICDE). New York: IEEE, 2023: 2303-2316. [119] SCHUSTER M, PALIWAL K K. Bidirectional recurrent neural networks[J]. IEEE Transactions on Signal Processing, 1997, 45(11): 2673-2681. doi: 10.1109/78.650093 [120] LEA C, VIDAL R, REITER A, et al. Temporal convolutional networks: A unified approach to action segmentation[C]//Springer. Proceedings of 12th European Conference on Computer Vision (ECCV). Munich: Springer, 2016: 47-54. [121] ZHANG X Y, JIN X Y, GOPALSWAMY K, et al. First de-trend then attend: Rethinking attention for time-series forecasting[C]//NeurIPS Foundation. Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2022: 1-11. [122] ZHOU T, MA Z Q, WEN Q S, et al. FEDformer: Frequency enhanced decomposed transformer for long-term series forecasting[C]//ACM. International Conference on Machine Learning. New York: ACM, 2022: 27268-27286. [123] ZHOU H Y, ZHANG S H, PENG J Q, et al. Informer: Beyond efficient transformer for long sequence time-series forecasting[C]//AAAI. Proceedings of the AAAI Confe-rence on Artificial Intelligence. Washington DC: AAAI Press, 2021: 11106-11115. [124] LIU Y, TU T G, ZHANG H R, et al. Transformer: Inverted transformers are effective for time series forecasting[C]//YOSHUA B, YANN L. International Conference on Learning Representations. Portland: OpenReview. net, 2024: 1-25. [125] LIU Y M, WU X, TANG Y, et al. Decomposition with feature attention and graph convolution network for traffic forecasting[J]. Knowledge-based Systems, 2024, 300: 112193. doi: 10.1016/j.knosys.2024.112193 [126] ZHOU H Y, ZHANG S H, PENG J Q, et al. Informer: Beyond efficient transformer for long sequence time-series forecasting[C]//AAAI. Proceedings of the AAAI Confe-rence on Artificial Intelligence. Washington DC: AAAI, 2021: 11106-11115. [127] LIU H C, DONG Z, JIANG R H, et al. Spatio-temporal adaptive embedding makes vanilla transformer SOTA for traffic forecasting[C]//ACM. Proceedings of the 32nd ACM International Conference on Information and Knowledge Management. New York: ACM, 2023: 4125-4129. [128] DAUPHIN Y N, FAN A, AULI M, et al. Language modeling with gated convolution networks[C]//ACM. International Conference on Machine Learning. New York: ACM, 2017: 933-941. [129] LIU Z M, WANG Y X, VAIDYA S, et al. KAN: Kolmogorov-arnold network[C]//YUE Y. International Conference on Learning Representations. Amherst: OpenReview. net, 2025: 1-50. [130] GAO Y, HU Z H, CHEN W A, et al. A revolutionary neural network architecture with interpretability and flexibility based on Kolmogorov-Arnold for solar radiation and temperature forecasting[J]. Applied Energy, 2025, 378: 124844. doi: 10.1016/j.apenergy.2024.124844 [131] LIVIERIS I E. C-KAN: A new approach for integrating convolutional layers with Kolmogorov-Arnold networks for time-series forecasting[J]. Mathematics, 2024, 12(19): 3022. doi: 10.3390/math12193022 [132] LIU X F, YANG Z L, GUO Y J, et al. A novel correlation feature self-assigned Kolmogorov-Arnold networks for multi-energy load forecasting in integrated energy systems[J]. Energy Conversion and Management, 2025, 325: 119388. doi: 10.1016/j.enconman.2024.119388 [133] GU A, DAO T. Mamba: Linear-time sequence modeling with selective state spaces[C]//YEJIN C, DANNY Z. Proceedings of 1st Conference on Language Modeling. Amherst: Open-Review. net, 2024: 1-36. [134] LIN W X, ZHANG Z, REN G, et al. MGCN: Mamba-integrated spatiotemporal graph convolutional network for long-term traffic forecasting[J]. Knowledge-based Systems, 2025, 309: 112875. doi: 10.1016/j.knosys.2024.112875 [135] MEHRABIAN A, WONG V W S. A-gamba: An adaptive graph-mamba model for traffic prediction in wireless cellular networks[J]. IEEE Wireless Communications Letters, 2025, 14(6): 1801-1805. doi: 10.1109/LWC.2025.3557313 [136] SHAO Z Q, WANG Z, YAO X S, et al. ST-MambaSync: Complement the power of Mamba and Transformer fusion for less computational cost in spatial-temporal traffic forecasting[J]. Information Fusion, 2025, 117: 102872. doi: 10.1016/j.inffus.2024.102872 [137] ZHOU J, CUI G Q, HU S D, et al. Graph neural networks: A review of methods and applications[J]. AI Open, 2020, 1: 57-81. doi: 10.1016/j.aiopen.2021.01.001 [138] 向仪, 冯强. 傅里叶正、余弦变换的加权卷积及其应用[J]. 浙江大学学报(理学版), 2023, 50(3): 266-272.XIANG Yi, FENG Qiang. Weighted convolution of the Fourier sine-cosine transform and its application[J]. Journal of Zhejiang University (Science Edition), 2023, 50(3): 266-272. [139] HAMMOND D K, VANDERGHEYNST P, GRIBONVAL R. Wavelets on graphs via spectral graph theory[J]. Applied and Computational Harmonic Analysis, 2011, 30(2): 129-150. doi: 10.1016/j.acha.2010.04.005 [140] HE M G, WEI Z W, HUANG Z F, et al. BernNet: Learning arbitrary graph spectral filters via Bernstein approximation[C]//NeurIPS Foundation. Conference on Neural Infor-mation Processing Systems. Cambridge: MIT Press, 2021: 1-16. [141] CHIEN E L, PENG J H, LI P, et al. Adaptive universal generalized PageRank graph neural network[C]//YOSHUA B, YANN L. International Conference on Learning Repre-sentations. Amherst: OpenReview. net, 2021: 1-24. [142] KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks[C]// YOSHUA B, YANN L. International Conference on Learning Representations. Amherst: OpenReview. net, 2017: 1-14. [143] GILMER J, SCHOENHOLZ S S, RILEY P F, et al. Neural message passing for quantum chemistry[C]//ACM. International Conference on Machine Learning. New York: ACM, 2017: 1263-1272. [144] HAMILTON W L, YING R, LESKOVEC J. Inductive representation learning on large graphs[C]//NeurIPS Foundation. Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2017: 1025-1035. [145] LOUIS-PASCAL A C X, QU M, TANG J. Continuous graph neural networks[C]//ACM. International Conference on Machine Learning. New York: ACM, 2020: 1-15. [146] KIPF T N, WELLING M. Variational graph auto-encoders[C]//NeurIPS Foundation. Conference on Neural Infor-mation Processing Systems. Cambridge: MIT Press, 2016: 1-3. [147] VELICKOVIC P, CUCURULL G, CASANOVA A, et al. Graph attention networks[C]//YOSHUA B, YANN L. International Conference on Learning Representations. Amherst: OpenReview. net, 2018: 1-12. [148] ZHANG J N, SHI X J, XIE J Y, et al. GaAN: Gated attention networks for learning on large and spatio-temporal graphs[C]//AUAI. Conference on Uncertainty in Artificial Intelligence. Berkeley: AUAI, 2018: 339-349. [149] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90. doi: 10.1145/3065386 [150] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//IEEE. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). New York: IEEE, 2016: 770-778. [151] SAMI A E H, BRYAN P, AMOL K, et al. MixHop: Higher-order graph convolutional architectures via sparsified neigh-borhood mixing[C]//ACM. International Conference on Machine Learning. New York: ACM, 2019: 21-29. [152] CHEN F W, PAN S R, JIANG J, et al. DAGCN: Dual attention graph convolutional networks[C]//IEEE. 2019 International Joint Conference on Neural Networks (IJCNN). New York: IEEE, 2019: 1-8. [153] HUANG G, LIU Z, VAN DER MAATEN L, et al. Densely connected convolutional networks[C]//IEEE. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). New York: IEEE, 2017: 2261-2269. [154] SUN Y F, JIANG X H, HU Y L, et al. Dual dynamic spatial-temporal graph convolution network for traffic prediction[J]. IEEE Transactions on Intelligent Trans-portation Systems, 2022, 23(12): 23680-23693. doi: 10.1109/TITS.2022.3208943 [155] SHANG C, CHEN J, BI J B. Discrete graph structure learning for forecasting multiple time series[C]//YOSHUA B, YANN L. International Conference on Learning Repre-sentations. Amherst: OpenReview. net, 2021: 1-14. [156] ZHENG C P, FAN X L, WANG C, et al. GMAN: A graph multi-attention network for traffic prediction[C]//AAAI. Proceedings of the AAAI Conference on Artificial Intel-ligence. Washington DC: AAAI Press, 2020: 1234-1241. [157] ZHANG Z C, LIN X, LI M, et al. A customized deep learning approach to integrate network-scale online traffic data imputation and prediction[J]. Transportation Research Part C: Emerging Technologies, 2021, 132: 103372. doi: 10.1016/j.trc.2021.103372 [158] WEN Q S, SUN L, YANG F, et al. Time series data augmentation for deep learning: A survey[C]//AAAI. Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. Washington DC: AAAI Press, 2021: 1-8. [159] ZHENG X, BAGLOEE S A, SARVI M. TRECK: Long-term traffic forecasting with contrastive representation learning[J]. IEEE Transactions on Intelligent Transportation Systems, 2024, 25(11): 16964-16977. doi: 10.1109/TITS.2024.3421328 [160] WANG X, AL-BASHABSHEH A, ZHAO C, et al. Smoothed noise contrastive mutual information neural estimation[J]. Journal of the Franklin Institute, 2023, 360(16): 12415-12435. doi: 10.1016/j.jfranklin.2023.08.047 [161] LAN Z X, REN Y L, YU H Y, et al. Hi-SCL: Fighting long-tailed challenges in trajectory prediction with hierarchical wave-semantic contrastive learning[J]. Transportation Research Part C: Emerging Technologies, 2024, 165: 104735. doi: 10.1016/j.trc.2024.104735 [162] QU Y S, RONG J, LI Z L, et al. ST-A-PGCL: Spatio-temporal adaptive periodical graph contrastive learning for traffic prediction under real scenarios[J]. Knowledge-based Systems, 2023, 272: 110591. doi: 10.1016/j.knosys.2023.110591 [163] YOU Y N, CHEN T L, SUI Y D, et al. Graph contrastive learning with augmentations[C]//NeurIPS Foundation. Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2020: 1-12. [164] 刘伟, 贾素玲. 基于图对比学习的稳健交通流量预测[J]. 北京航空航天大学学报, 2024, 50(1): 122-133.LIU Wei, JIA Su-ling. Robust traffic flow prediction based on graph contrastive learning[J]. Journal of Beijing University of Aeronautics and Astronautics, 2024, 50(1): 122-133. [165] JIA R, GAO K, LIU Y, et al. I-CLTP: Integrated contrastive learning with transformer framework for traffic state prediction and network-wide analysis[J]. Transportation Research Part C: Emerging Technologies, 2025, 171: 104979. doi: 10.1016/j.trc.2024.104979 [166] JIN Y L, CHEN K, YANG Q. Selective cross-city transfer learning for traffic prediction via source city region re-weighting[C]//ACM. Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York: ACM, 2022: 734-741. [167] 赖培源, 李程, 王增辉, 等. 基于图提示微调的交通流量预测[J]. 计算机研究与发展, 2024, 61(8): 2020-2029.LAI Pei-yuan, LI Cheng, WANG Zeng-hui, et al. Traffic flow prediction based on graph prompt-finetuning[J]. Journal of Computer Research and Development, 2024, 61(8): 2020-2029. [168] QU Y S, LI Z L, ZHAO X H, et al. Towards real-world traffic prediction and data imputation: A multi-task pretraining and fine-tuning approach[J]. Information Sciences, 2024, 657: 119972. doi: 10.1016/j.ins.2023.119972 [169] SHAO Z Z, ZHANG Z, WANG F, et al. Pre-training enhanced spatial-temporal graph neural network for multivariate time series forecasting[C]//ACM. Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York: ACM, 2022: 1567-1577. [170] LI C, LIU W, YANG H. Deep causal inference for understanding the impact of meteorological variations on traffic[J]. Transportation Research Part C: Emerging Technologies, 2024, 165: 104744. doi: 10.1016/j.trc.2024.104744 [171] LIU J M, LIN H, WANG X D, et al. Reliable trajectory prediction in scene fusion based on spatio-temporal structure causal model[J]. Information Fusion, 2024, 107: 102309. doi: 10.1016/j.inffus.2024.102309 [172] HU J J, BAI J, YANG J Y, et al. Crash risk prediction using sparse collision data: Granger causal inference and graph convolutional network approaches[J]. Expert Systems with Applications, 2025, 259: 125315. doi: 10.1016/j.eswa.2024.125315 [173] YIN J, LI B. Long-short-term expert attention neural networks for traffic flow prediction[C]//Springer. Advanced Intel-ligent Computing Technology and Applications. Munich: Springer, 2024: 3-14. [174] 李舒豪. ST-MoE: 用于交通预测去偏的时空混合专家框架[D]. 广州: 广州大学, 2023.LI Shu-hao. ST-MoE: A spatio-temporal mixed expert framework for traffic forecasting depolarization[D]. Guangzhou: Guangzhou University, 2023. [175] JIANG W Z, HAN J D, LIU H, et al. Interpretable cascading mixture-of-experts for urban traffic congestion prediction[C]//ACM. Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York: ACM, 2024: 5206-5217. [176] GUO S N, LIN Y F, GONG L T, et al. Self-supervised spatial-temporal bottleneck attentive network for efficient long-term traffic forecasting[C]//IEEE. 2023 IEEE 39th International Conference on Data Engineering (ICDE). New York: IEEE, 2023: 1585-1596. [177] ZHANG D R, YAN J N, POLAT K, et al. Multimodal joint prediction of traffic spatial-temporal data with graph sparse attention mechanism and bidirectional temporal convolutional network[J]. Advanced Engineering Informatics, 2024, 62: 102533. doi: 10.1016/j.aei.2024.102533.2022.1096186 [178] FENG S F, FENG X X, XU L X, et al. BTD-GTAN: Federated traffic flow prediction with multimodal feature fusion considering anomalies[C]//IEEE. 2024 9th Inter-national Conference on Computer and Communication Systems (ICCCS). New York: IEEE, 2024: 462-467. [179] WANG B, QIN A K, SHAFIEI S, et al. Training physics-informed neural networks via multi-task optimization for traffic density prediction[C]//IEEE. 2023 International Joint Conference on Neural Networks (IJCNN). New York: IEEE, 2023: 1-7. [180] DENG P, ZHAO Y, LIU J T, et al. Spatio-temporal neural structural causal models for bike flow prediction[J]//AAAI. Proceedings of the AAAI Conference on Artificial Intel-ligence. Washington DC: AAAI Press, 2023: 4242-4249. [181] WANG H J, CHEN J Y, PAN T, et al. Easy begun is half done: Spatial-temporal graph modeling with ST-curriculum dropout[C]//AAAI. Proceedings of the AAAI Conference on Artificial Intelligence. Washington DC: AAAI Press, 2023: 4668-4675. [182] GUO H, MEESE C, LI W X, et al. B2SFL: A bi-level blockchained architecture for secure federated learning-based traffic prediction[J]. IEEE Transactions on Services Computing, 2023, 16(6): 4360-4374. doi: 10.1109/TSC.2023.3318990 [183] LU B, GAN X Y, ZHANG W N, et al. Spatio-temporal graph few-shot learning with cross-city knowledge transfer[C]//ACM. Proceedings of the 28th ACM SIGKDD Confe-rence on Knowledge Discovery and Data Mining. New York: ACM, 2022: 1162-1172. [184] GUO X S, ZHANG Q M, JIANG J Y, et al. Towards explainable traffic flow prediction with large language models[J]. Communications in Transportation Research, 2024, 4: 100150. doi: 10.1016/j.commtr.2024.100150 [185] YUAN Y, DING J T, FENG J, et al. UniST: A prompt-empowered universal model for urban spatio-temporal prediction[C]//ACM. Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York: ACM, 2024: 4095-4106. [186] 肖建力, 邱雪, 张扬, 等. 交通大模型综述[J]. 交通运输工程学报, 2025, 25(1): 8-28. doi: 10.19818/j.cnki.1671-1637.2025.01.002XIAO Jian-li, QIU Xue, ZHANG Yang, et al. Review on large language models in transportation[J]. Journal of Traffic and Transportation Engineering, 2025, 25(1): 8-28. doi: 10.19818/j.cnki.1671-1637.2025.01.002 [187] 董瀚萱. 数据丢失情况下高速路网交通状态预测方法研究[D]. 南京: 东南大学, 2022.DONG Han-xuan. Research on traffic state prediction method of expressway network under data loss[D]. Nanjing: Southeast University, 2022. [188] 赵永梅, 董云卫. 低秩张量补全的时空交通数据预测[J]. 交通运输工程学报, 2024, 24(4): 243-258. doi: 10.19818/j.cnki.1671-1637.2024.04.018ZHAO Yong-mei, DONG Yun-wei. Spatio-temporal traffic data prediction based on low-rank tensor completion[J]. Journal of Traffic and Transportation Engineering, 2024, 24(4): 243-258. doi: 10.19818/j.cnki.1671-1637.2024.04.018 -
图(14) / 表(4)
计量
- 文章访问数: 26
- HTML全文浏览量: 10
- PDF下载量: 5
- 被引次数: 0
