Cross-modal missing time-series imputation using dense spatio-temporal transformer nets.

IF 2.6 4区 工程技术 Q1 Mathematics
Xusheng Qian, Teng Zhang, Meng Miao, Gaojun Xu, Xuancheng Zhang, Wenwu Yu, Duxin Chen
{"title":"Cross-modal missing time-series imputation using dense spatio-temporal transformer nets.","authors":"Xusheng Qian, Teng Zhang, Meng Miao, Gaojun Xu, Xuancheng Zhang, Wenwu Yu, Duxin Chen","doi":"10.3934/mbe.2024220","DOIUrl":null,"url":null,"abstract":"<p><p>Due to irregular sampling or device failure, the data collected from sensor network has missing value, that is, missing time-series data occurs. To address this issue, many methods have been proposed to impute random or non-random missing data. However, the imputation accuracy of these methods are not accurate enough to be applied, especially in the case of complete data missing (CDM). Thus, we propose a cross-modal method to impute time-series missing data by dense spatio-temporal transformer nets (DSTTN). This model embeds spatial modal data into time-series data by stacked spatio-temporal transformer blocks and deployment of dense connections. It adopts cross-modal constraints, a graph Laplacian regularization term, to optimize model parameters. When the model is trained, it recovers missing data finally by an end-to-end imputation pipeline. Various baseline models are compared by sufficient experiments. Based on the experimental results, it is verified that DSTTN achieves state-of-the-art imputation performance in the cases of random and non-random missing. Especially, the proposed method provides a new solution to the CDM problem.</p>","PeriodicalId":49870,"journal":{"name":"Mathematical Biosciences and Engineering","volume":"21 4","pages":"4989-5006"},"PeriodicalIF":2.6000,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Mathematical Biosciences and Engineering","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.3934/mbe.2024220","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"Mathematics","Score":null,"Total":0}
引用次数: 0

Abstract

Due to irregular sampling or device failure, the data collected from sensor network has missing value, that is, missing time-series data occurs. To address this issue, many methods have been proposed to impute random or non-random missing data. However, the imputation accuracy of these methods are not accurate enough to be applied, especially in the case of complete data missing (CDM). Thus, we propose a cross-modal method to impute time-series missing data by dense spatio-temporal transformer nets (DSTTN). This model embeds spatial modal data into time-series data by stacked spatio-temporal transformer blocks and deployment of dense connections. It adopts cross-modal constraints, a graph Laplacian regularization term, to optimize model parameters. When the model is trained, it recovers missing data finally by an end-to-end imputation pipeline. Various baseline models are compared by sufficient experiments. Based on the experimental results, it is verified that DSTTN achieves state-of-the-art imputation performance in the cases of random and non-random missing. Especially, the proposed method provides a new solution to the CDM problem.

利用密集时空变换网进行跨模态缺失时间序列估算。
由于不规则采样或设备故障,从传感器网络收集到的数据会出现缺失值,即时间序列数据缺失。为解决这一问题,人们提出了许多方法来估算随机或非随机缺失数据。然而,这些方法的估算精度不够准确,尤其是在数据完全缺失(CDM)的情况下。因此,我们提出了一种跨模态方法,通过密集时空变换网(DSTTN)来估算时间序列缺失数据。该模型通过堆叠时空变换块和部署密集连接,将空间模态数据嵌入时间序列数据。它采用跨模态约束、图拉普拉斯正则化项来优化模型参数。模型训练完成后,通过端到端的估算管道最终恢复缺失数据。通过充分的实验对各种基线模型进行了比较。根据实验结果,验证了 DSTTN 在随机和非随机缺失的情况下都达到了最先进的估算性能。特别是,所提出的方法为 CDM 问题提供了一种新的解决方案。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Mathematical Biosciences and Engineering
Mathematical Biosciences and Engineering 工程技术-数学跨学科应用
CiteScore
3.90
自引率
7.70%
发文量
586
审稿时长
>12 weeks
期刊介绍: Mathematical Biosciences and Engineering (MBE) is an interdisciplinary Open Access journal promoting cutting-edge research, technology transfer and knowledge translation about complex data and information processing. MBE publishes Research articles (long and original research); Communications (short and novel research); Expository papers; Technology Transfer and Knowledge Translation reports (description of new technologies and products); Announcements and Industrial Progress and News (announcements and even advertisement, including major conferences).
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信