MulTiSA 2025

2nd Workshop on Multivariate Time Series Analytics

in conjunction with ICDE’25 | 19th May 2025, Hong Kong, China.

MulTiSA 2025

About

Contemporary algorithms for time series management predominantly handle univariate time series. Modern data sources frequently generate richer, multivariate time series. Examples include sensors monitoring multiple variables (e.g., temperature, wind, rainfall), financial time series (bid/ask prices, volume), and data from scientific and medical equipment. Currently, only very few algorithms address the management, analysis, and extraction of insights from such multivariate data. Moreover, existing work is tailored to specific needs. Foundational functionalities that have propelled advancements in univariate time series analysis, e.g., indexing, cannot be trivially extended to the multivariate case. This limitation significantly restricts existing efforts for analyzing multivariate time series.

This workshop will bring together researchers and practitioners working with multivariate time series, to present and discuss open problems and solutions, and to foster collaborations. Industry will participate for presenting requirements and current approaches, and to reach out to the ICDE community. Researchers will share their novel and ongoing work. The workshop will feature: (a) paper presentations (short papers and demos up to 4 pages, regular papers up to 8 pages – excluding references), (b) invited talks from industry and domain experts, (c) panel discussion, and time for fostering collaborations.

All accepted papers will be published by IEEE, in the ICDE workshop proceedings – alongside the ICDE conference proceedings.

[ Go to Top ]

Topics of interest

The topics of interest include (but are not limited to):

[ Go to Top ]

Submission Guidelines

The workshop will accept regular papers (up to 8 pages, excluding references) and short papers describing work in progress, demos, vision/outrageous ideas (up to 4 pages, excluding references). All submissions must be prepared in accordance with the IEEE template available here. The workshop follows the same rules of Conflicts of Interest (COI) as ICDE 2025. The following are the page limits (excluding references):

Regular papers: 8 pages
Short papers: 4 pages

All submissions (in PDF format) should be submitted to Microsoft CMT.

[ Go to Top ]

Important Dates

All deadlines are 11:59PM AoE.

Submission deadline: February 15, 2025
Notifications: March 10, 2025
Camera-ready deadline: March 24, 2025
Workshop date: May 19, 2025
Registration deadline for accepted papers to appear in the proceedings: April 15, 2025
[ Go to Top ]

Program

[ Go to Top ]

Panel Chair

Paul

Paul Boniol

Inria, France

Bio: Paul Boniol is a Researcher at Inria, in the VALDA project-team. Previously, he worked at ENS Paris-Saclay, Université Paris Cité, EDF Research, and Ecole Polytechnique. His research interests lie between data management, machine learning, and time-series analysis. His Ph.D. focused on subsequence anomaly detection and time-series classification, and he won several PhD awards, including the Paul Caseau Prize, supported by the Academy of Sciences of France. His work has been published in the top data management and data mining venues.

Link to CV

Panelists

Bin

Bin Yang

East China Normal University

Bio: Bin Yang is Chair Professor at School of Data Science and Engineering, East China Normal University. He has been a Full Professor in Department of Computer Science, Aalborg University, Denmark. Previously, he was at Max-Planck-Institut für Informatik, Germany, and at Aarhus University, Denmark. He obtained his Ph.D. degree from Fudan University. His research interests cover artificial intelligence and data governance, with a focus on enabling data driven decision making with time series and spatio-temporal data. Recently, much of his research concerns the AGREE principles—Automation, Generalization, Robustness, Explainability, and Efficiency, on a variety of tasks, e.g., forecasting, outlier detection, classification, ranking, searching, and decision making.

Byron

Byron Choi

Hong Kong Baptist University

Bio: Byron Choi is the Associate Head (Teaching and Learning) and a Full Professor at the Department of Computer Science at Hong Kong Baptist University. His research focuses on database systems, particularly time series analysis, visual graph querying, and graph database security. He has served as a reviewer of ACM SIGMOD, PVLDB, IEEE ICDE, and ACM CIKM. He was awarded one of the best reviewers of ACM SIGMOD 2023 and a distinguished program committee member from ACM SIGMOD 2021. He received the distinguished reviewer award from PVLDB 2019. He is an Associate Editor of Distributed and Parallel Databases (DAPD). He is a Fellow of the Hong Kong Institution of Engineers (HKIE), the Information discipline.

Mourad

Mourad Khayati

University of Fribourg, Switzerland

Bio: Mourad Khayati is a senior researcher and lecturer at the Department of Computer Science of the University of Fribourg, Switzerland. He obtained his PhD from the University of Zurich, Switzerland, under the supervision of Prof. Michael Böhlen. His research interests include Time Series analytics, data repair, missing values imputation, and time series database systems. His imputation benchmark won the VLDB 2020 Most Reproducible Paper Award and his ORBITS system is featured in the VLDB reproducibility highlights. He recently served as an area chair for the Knowledge Data Discovery (KDD) Conference 2025 and the ACM International Conference on Information and Knowledge Management (CIKM) 2020. He regularly serves as a reviewer for various DB and data mining journals and conferences, including, the VLDB Journal, TKDE, KDD, and EDBT.

Chen

Chen Wang

Tsinghua University, China

Bio: Chen Wang is a Research Associate Professor and CTO at the National Engineering Research Center for Big Data Software, Tsinghua University, where he also serves as Director of the Big Data Research Department at the University’s Energy Internet Research Institute. Prior to joining Tsinghua University, he was a Research Staff Member and Senior Manager in the Information Management Research Department at IBM Research China. His research focuses on database systems, data governance, time-series data management, and industrial big data applications. A founding member and PMC member of the Apache IoTDB project, Prof. Wang currently also holds the position of Chief Scientist at Timecho Inc., the commercial entity driving IoTDB’s industry adoption. Additional academic profiles can be accessed via his homepage: https://wc81.github.io

[ Go to Top ]

Keynote Talks

Keynote 1: Multivariate Time Series Management and Analysis: An Apache IoTDB Solution

Shaoxu Song, Tsinghua University

Abstract: Multivariate time series are now prevalent, especially in the IoT scenarios, for monitoring massive devices and supporting decision making. In this talk, we will first introduce some typical scenarios of managing and analyzing multivariate time series, and give an overview on how Apache IoTDB can meet such user requirements. Our solution ranges from storage, compression and curation for time series management, to statistical analysis, machine learning and pre-trained models integrated with the database. We will also share several use cases of IoTDB in real-world applications, and indicate some future directions to advance the study further.

Bio: Shaoxu Song is an Associate Professor in the School of Software at Tsinghua University. His research interests include timeseries database, data quality and data mining. He has published more than 80 research papers in top journals and conferences in the area, such as TODS, VLDBJ, TKDE, SIGMOD, KDD, VLDB, and ICDE. Dr. Song served in the editorial (review) boards of PVLDB, JDIQ, JCST and ESWA, the PC Vice-Co-Chair of IEEE BigData 2022/2025 and Area Chair of ACM KDD 2025. He received Outstanding/Distinguished Reviewer awards from KDD 2025, ICDE 2024, VLDB 2019 and CIKM 2017. He is a Senior Member of IEEE.

Keynote 2: Time Series Foundation Models
Bin

Bin Yang

East China Normal University

Abstract: As part of the continued digitalization of processes throughout society, increasingly large volumes of time series are available, ranging from the scientific and medical domains to the industrial and environmental domains. In this talk, we focus on decision making with time series data, e.g., enabling greener and more efficient transportation based on traffic time series forecasting. We introduce our research paradigm of “data-governance-analytics-decision,” and the AGREE principles (Automation, Generalization, Robustness, Explainability, and Efficiency). Specifically, we focus on our recent works on time series foundation models for forecasting, anomaly detection, and classification. Finally, we cover time series benchmarking.

Bio: Bin Yang is Chair Professor at School of Data Science and Engineering, East China Normal University. He has been a Full Professor in Department of Computer Science, Aalborg University, Denmark. Previously, he was at Max-Planck-Institut für Informatik, Germany, and at Aarhus University, Denmark. He obtained his Ph.D. degree from Fudan University. His research interests cover artificial intelligence and data governance, with a focus on enabling data driven decision making with time series and spatio-temporal data. Recently, much of his research concerns the AGREE principles—Automation, Generalization, Robustness, Explainability, and Efficiency, on a variety of tasks, e.g., forecasting, outlier detection, classification, ranking, searching, and decision making.

[ Go to Top ]

Organizers

[ Go to Top ]

Program Committee

[ Go to Top ]

Web and publicity chair

[ Go to Top ]

Past Workshops

[ Go to Top ]