中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 地理科学与资源研究所
  3. 中国科学院地理科学与资源研究所
  4. 资源与环境信息系统国家重点实验室
  5. 外文论文
Heterogeneity-Guided Tensor Decomposition for Traffic Flow Imputation

文献类型：期刊论文

作者Luo, Xiaoyue5,6; Cheng, Shifen5,6; Yang, Yitao4; Wang, Peixiao5,6; Guo, Shengmin1; Lu, Feng2,3,5,6
刊名IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
出版日期2026-02-06
卷号N/A
关键词Traffic flow imputation spatiotemporal heterogeneity representation learning tensor decomposition self-supervised learning
ISSN号1524-9050
DOI10.1109/TITS.2026.3658186
产权排序1
文献子类Article ; Early Access
英文摘要Tensor decomposition has been widely recognized as a promising approach for addressing missing data in traffic flow analysis. Recent studies have demonstrated its effectiveness in capturing complex spatiotemporal patterns and improving imputation accuracy. However, two challenges still remain: i)The spatiotemporal heterogeneity at missing locations is difficult to identify and accurately represent, preventing the model from optimizing based on the true data distribution; ii) The lack of a dynamic feedback mechanism hinders the model to adaptively optimize its spatiotemporal representation based on imputation results, thus limiting the generalization and robustness. This paper proposes a heterogeneity-guided tensor decomposition method (Hg-TD) to address these problems. First, we design a spatiotemporal-decoupled masking and encoding strategy, leveraging contrastive learning based on reconstruction loss to generate a dynamic spatial weight matrix, thereby enhancing the representation of spatiotemporal heterogeneity at missing locations. Second, we introduce a dynamic feedback-driven self-supervised learning mechanism that uses imputation results as feedback to dynamically optimize both representation learning and tensor decomposition, allowing the model to adaptively adjust spatiotemporal constraints, thereby improving imputation performance under sparse and non-stationary data conditions. Experimental results on four traffic flow datasets from urban areas demonstrate that Hg-TD consistently surpasses fifteen baseline models across various data missing patterns, and achieves a favorable balance between computational complexity and imputation performance. The improvement relative to the baselines is particularly pronounced under conditions of severe data missingness. Codes and datasets are available at https://figshare.com/s/17b6beeb8e8bad5e3ee7
URL标识查看原文
WOS关键词NEURAL-NETWORKS ; RECOGNITION ; MODEL
WOS研究方向Engineering ; Transportation
语种英语
WOS记录号WOS:001682809900001
出版者IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
源URL[http://ir.igsnrr.ac.cn/handle/311030/220996]  
专题资源与环境信息系统国家重点实验室_外文论文
通讯作者Cheng, Shifen
作者单位1.Beijing PalmGo Infotech Co Ltd, Beijing 100094, Peoples R China;
2.Jiangsu Ctr Collaborat Innovat Geog Informat Resou, Nanjing 210023, Peoples R China
3.Fuzhou Univ, Acad Digital China, Fuzhou 350003, Peoples R China;
4.Univ Leeds, Sch Geog, Leeds LS2 9JT, England;
5.Univ Chinese Acad Sci, Beijing 100049, Peoples R China;
6.Chinese Acad Sci, State Key Lab Resources & Environm Informat Syst, Inst Geog Sci & Nat Resources Res, Beijing 100101, Peoples R China;
推荐引用方式
GB/T 7714		 Luo, Xiaoyue,Cheng, Shifen,Yang, Yitao,et al. Heterogeneity-Guided Tensor Decomposition for Traffic Flow Imputation[J]. IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS,2026,N/A.
APA Luo, Xiaoyue,Cheng, Shifen,Yang, Yitao,Wang, Peixiao,Guo, Shengmin,&Lu, Feng.(2026).Heterogeneity-Guided Tensor Decomposition for Traffic Flow Imputation.IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS,N/A.
MLA Luo, Xiaoyue,et al."Heterogeneity-Guided Tensor Decomposition for Traffic Flow Imputation".IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS N/A(2026).

入库方式： OAI收割

来源：地理科学与资源研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。