CL-DGCN: contrastive learning based deeper graph convolutional network for traffic flow data prediction

doi:10.1016/j.tre.2025.104345

	CL-DGCN: contrastive learning based deeper graph convolutional network for traffic flow data prediction
	Zhang, Enwei 1; Lv, Zhiqiang 1,2; Cheng, Zesheng 1; Ke, Jintao 3
	2025-11-01
发表期刊	TRANSPORTATION RESEARCH PART E-LOGISTICS AND TRANSPORTATION REVIEW
ISSN	1366-5545
卷号	203 页码:18
摘要	Accurate and efficient traffic prediction helps to establish multimodal transportation systems and improve the travelling experience in daily life. Currently the mainstream traffic prediction methods are implemented based on Graph Convolutional Network (GCN), superimposing GCN layers can obtain better prediction results, but excessive superimposition will lead to the oversmooth problem, this paper proposes CL-DGCN to overcome this problem, which obtains the representations of the features through contrastive learning, and uses the improved message aggregation function to overcome the over-smooth problem. In this study, the CL-DGCN model is experimented on four domestic and international open-source, real datasets (PEMSBAY, METRLA, BEIJING and SZ-TAXI), and CL-DGCN achieves optimal or sub-optimal results in most time-step predictions, and reduces the composite error by more than 10 % compared to the baseline model, which well illustrates that the CL-DGCN model possesses more accurate prediction results.
关键词	Multimodal transportation Traffic flow prediction Graph convolutional network Hyperaggregation function
DOI	10.1016/j.tre.2025.104345
收录类别	SCI
语种	英语
资助项目	Key Technology Research and Development Program of Shandong[2025CXGC010108] ; Shandong Province Natural Science Foundation[ZR2024MG034] ; Shandong Province Natural Science Foundation[ZR2024MF144] ; Shandong Province Natural Science Foundation[ZR2024MF142]
WOS研究方向	Business & Economics ; Engineering ; Operations Research & Management Science ; Transportation
WOS类目	Economics ; Engineering, Civil ; Operations Research & Management Science ; Transportation ; Transportation Science & Technology
WOS记录号	WOS:001550879900002
出版者	PERGAMON-ELSEVIER SCIENCE LTD
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/41771
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Cheng, Zesheng
作者单位	1.Qingdao Univ, Coll Comp Sci & Technol, Qingdao, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 3.Univ Hong Kong, Dept Civil Engn, Hong Kong, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Enwei,Lv, Zhiqiang,Cheng, Zesheng,et al. CL-DGCN: contrastive learning based deeper graph convolutional network for traffic flow data prediction[J]. TRANSPORTATION RESEARCH PART E-LOGISTICS AND TRANSPORTATION REVIEW,2025,203:18.
APA	Zhang, Enwei,Lv, Zhiqiang,Cheng, Zesheng,&Ke, Jintao.(2025).CL-DGCN: contrastive learning based deeper graph convolutional network for traffic flow data prediction.TRANSPORTATION RESEARCH PART E-LOGISTICS AND TRANSPORTATION REVIEW,203,18.
MLA	Zhang, Enwei,et al."CL-DGCN: contrastive learning based deeper graph convolutional network for traffic flow data prediction".TRANSPORTATION RESEARCH PART E-LOGISTICS AND TRANSPORTATION REVIEW 203(2025):18.