App. Res. of Comp. Accepted Paper
Vol. 43, 2026 No. 9

Missingness-aware dual-branch attention network for multivariate time series imputation

Zhang Feng
Li Wenli
Jiang Hongyang
Dong Chunru
Zhu Jie
Hebei Key Laboratory of Machine Learning and Computational Intelligence, College of Mathematics & Information Science, Hebei University, Baoding Hebei 071002, China
DOI: 10.19734/j.issn.1001-3695.2025.12.0537

Abstract

To address the prevalent missing data problem in multivariate time series, existing attention-based imputation methods often struggle to simultaneously capture three essential types of information: the temporal dynamics of each individual variable, the correlations among variables, and the trends shared across all variables. In addition, these methods insufficiently exploit latent information embedded in missing patterns, leading to biased imputation results. To overcome these limitations, a Missingness-Aware Dual-branch Attention Network (MDual-AttNet) was developed. Specifically, MDual-AttNet employed a gate-based fusion mechanism at the embedding layer to adaptively integrate observations and missing patterns, generating a missingness-aware representation. The network then processed this representation using two attention-based branches. One branch captured the temporal dependencies of each variable and correlations between variables through block-level self-attention and graph attention. The other branch modeled the shared dynamics across all variables using point-level self-attention. Additionally, MDual-AttNet introduced a mask prediction auxiliary task to explicitly guide the learning of missing patterns and optimize imputation performance. Empirical results on four real-world datasets from medical, environmental, and energy domains showed that MDual-AttNet achieved an average reduction of 16% in Mean Absolute Error (MAE) compared with representative baseline methods. Moreover, the ROC-AUC metric improved by 1.0% in downstream classification tasks when using the imputed data, indicating the effectiveness of MDual-AttNet in practical applications.

Foundation Support

国家重点研发计划(2022YFE0196100)
河北省自然科学基金项目(F2018201115,F2022511001)
河北大学高层次人才科研启动项目(521000981095)
河北省创新能力提升计划-科技平台项目(22567623H)

Publish Information

DOI: 10.19734/j.issn.1001-3695.2025.12.0537
Publish at: Application Research of Computers Accepted Paper, Vol. 43, 2026 No. 9

Publish History

[2026-05-25] Accepted Paper

Cite This Article

张峰, 李文丽, 蒋宏阳, 等. 基于缺失感知与双分支注意力的多变量时间序列插补方法 [J]. 计算机应用研究, 2026, 43 (9). (2026-06-02). https://doi.org/10.19734/j.issn.1001-3695.2025.12.0537. (Zhang Feng, Li Wenli, Jiang Hongyang, et al. Missingness-aware dual-branch attention network for multivariate time series imputation [J]. Application Research of Computers, 2026, 43 (9). (2026-06-02). https://doi.org/10.19734/j.issn.1001-3695.2025.12.0537. )

About the Journal

  • Application Research of Computers Monthly Journal
  • Journal ID ISSN 1001-3695
    CN  51-1196/TP

Application Research of Computers, founded in 1984, is an academic journal of computing technology sponsored by Sichuan Institute of Computer Sciences under the Science and Technology Department of Sichuan Province.

Aiming at the urgently needed cutting-edge technology in this discipline, Application Research of Computers reflects the mainstream technology, hot technology and the latest development trend of computer application research at home and abroad in a timely manner. The main contents of the journal include high-level academic papers in this discipline, the latest scientific research results and major application results. The contents of the columns involve new theories of computer discipline, basic computer theory, algorithm theory research, algorithm design and analysis, blockchain technology, system software and software engineering technology, pattern recognition and artificial intelligence, architecture, advanced computing, parallel processing, database technology, computer network and communication technology, information security technology, computer image graphics and its latest hot application technology.

Application Research of Computers has many high-level readers and authors, and its readers are mainly senior and middle-level researchers and engineers engaged in the field of computer science, as well as teachers and students majoring in computer science and related majors in colleges and universities. Over the years, the total citation frequency and Web download rate of Application Research of Computers have been ranked among the top of similar academic journals in this discipline, and the academic papers published are highly popular among the readers for their novelty, academics, foresight, orientation and practicality.

Indexed & Evaluation

  • The Second National Periodical Award 100 Key Journals
  • Double Effect Journal of China Journal Formation
  • the Core Journal of China (Peking University 2023 Edition)
  • the Core Journal for Science
  • Chinese Science Citation Database (CSCD) Source Journals
  • RCCSE Chinese Core Academic Journals
  • Journal of China Computer Federation
  • 2020-2022 The World Journal Clout Index (WJCI) Report of Scientific and Technological Periodicals
  • Full-text Source Journal of China Science and Technology Periodicals Database
  • Source Journal of China Academic Journals Comprehensive Evaluation Database
  • Source Journals of China Academic Journals (CD-ROM Version), China Journal Network
  • 2017-2019 China Outstanding Academic Journals with International Influence (Natural Science and Engineering Technology)
  • Source Journal of Top Academic Papers (F5000) Program of China's Excellent Science and Technology Journals
  • Source Journal of China Engineering Technology Electronic Information Network and Electronic Technology Literature Database
  • Source Journal of British Science Digest (INSPEC)
  • Japan Science and Technology Agency (JST) Source Journal
  • Russian Journal of Abstracts (AJ, VINITI) Source Journals
  • Full-text Journal of EBSCO, USA
  • Cambridge Scientific Abstracts (Natural Sciences) (CSA(NS)) core journals
  • Poland Copernicus Index (IC)
  • Ulrichsweb (USA)