research-article

Latent Time-Series Motifs

Authors:

Josif Grabocka,

Nicolas Schilling,

Lars Schmidt-ThiemeAuthors Info & Claims

ACM Transactions on Knowledge Discovery from Data (TKDD), Volume 11, Issue 1

Article No.: 6, Pages 1 - 20

https://doi.org/10.1145/2940329

Published: 20 July 2016 Publication History

Abstract

Motifs are the most repetitive/frequent patterns of a time-series. The discovery of motifs is crucial for practitioners in order to understand and interpret the phenomena occurring in sequential data. Currently, motifs are searched among series sub-sequences, aiming at selecting the most frequently occurring ones. Search-based methods, which try out series sub-sequence as motif candidates, are currently believed to be the best methods in finding the most frequent patterns.

However, this paper proposes an entirely new perspective in finding motifs. We demonstrate that searching is non-optimal since the domain of motifs is restricted, and instead we propose a principled optimization approach able to find optimal motifs. We treat the occurrence frequency as a function and time-series motifs as its parameters, therefore we learn the optimal motifs that maximize the frequency function. In contrast to searching, our method is able to discover the most repetitive patterns (hence optimal), even in cases where they do not explicitly occur as sub-sequences. Experiments on several real-life time-series datasets show that the motifs found by our method are highly more frequent than the ones found through searching, for exactly the same distance threshold.

References

[1]

André E. X. Brown, Eviatar I. Yemini, Laura J. Grundy, Tadas Jucikas, and William R. Schafer. 2013. A dictionary of behavioral motifs reveals clusters of genes affecting Caenorhabditis elegans locomotion. Proc. Natl. Acad. Sci. 110, 2 (2013), 791--796.

[2]

Jeremy Buhler and Martin Tompa. 2001. Finding motifs using random projections. In Proceedings of the 5th Annual International Conference on Computational Biology (RECOMB’01). ACM, New York, NY, 69--76.

Digital Library

[3]

N. Castro and P. Azevedo. 2010. Multiresolution motif discovery in time series. In Proceedings of the SIAM International Conference on Data Mining (SDM’10), Columbus, Ohio. SIAM, 665--676.

[4]

N. Castro and P. Azevedo. 2011. Time series motifs statistical significance. In Proceedings of the SIAM International Conference on Data Mining (SDM’11), Mesa, Arizona. SIAM, 687--698.

[5]

Nuno C. Castro and Paulo J. Azevedo. 2012. Significant motifs in time series. Stat. Anal. Data Min. 5, 1 (Feb. 2012), 35--53.

[6]

Joe Catalano, Tom Armstrong, and Tim Oates. 2006. Discovering patterns in real-valued time series. In Proceedings of the Knowledge Discovery in Databases (PKDD’06), Johannes Frnkranz, Tobias Scheffer, and Myra Spiliopoulou (Eds.). Lecture Notes in Computer Science, Vol. 4213. Springer Berlin, 462--469.

[7]

Bill Chiu, Eamonn Keogh, and Stefano Lonardi. 2003. Probabilistic discovery of time series motifs. In Proceedings of the 9th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD’03). ACM, New York, NY, 493--498.

Digital Library

[8]

Thomas H. Cormen, Clifford Stein, Ronald L. Rivest, and Charles E. Leiserson. 2001. Introduction to Algorithms (2nd ed.). McGraw-Hill Higher Education.

Digital Library

[9]

John Duchi, Elad Hazan, and Yoram Singer. 2011. Adaptive subgradient methods for online learning and stochastic optimization. J. Mach. Learn. Res. 12 (July 2011), 2121--2159.

Digital Library

[10]

Pedro G. Ferreira, Paulo J. Azevedo, Candida G. Silva, and Rui M. M. Brito. 2006. Mining approximate motifs in time series. In Discovery Science, Ljupco Todorovski, Nada Lavrac, and Klaus P. Jantke (Eds.). Lecture Notes in Computer Science, Vol. 4265. Springer Berlin, 89--101. 11893318_12

Digital Library

[11]

A. L. Goldberger, L. A. N. Amaral, L. Glass, J. M. Hausdorff, P. Ch. Ivanov, R. G. Mark, J. E. Mietus, G. B. Moody, C.-K. Peng, and H. E. Stanley. 2000 (June 13). PhysioBank, PhysioToolkit, and PhysioNet: Components of a New Research Resource for Complex Physiologic Signals. Circulation 101, 23 (2000 (June 13)), e215--e220.

[12]

Hoang Thanh Lam, Ninh Dang Pham, and Toon Calders. 2011. Online Discovery of Top-k Similar Motifs in Time Series Data. Chapter 86, 1004--1015.

[13]

Yuan Li and Jessica Lin. 2010. Approximate variable-length time series motif discovery using grammar inference. In Proceedings of the 10th International Workshop on Multimedia Data Mining (MDMKDD’10). ACM, New York, NY, Article 10, 9 pages.

Digital Library

[14]

Yuan Li, Jessica Lin, and Tim Oates. 2012. Visualizing variable-length time series motifs. In Proceedings of the 12th SIAM International Conference on Data Mining, Anaheim, California, April 26--28. SIAM / Omnipress, 895--906.

[15]

Zheng Liu, JeffreyXu Yu, Xuemin Lin, Hongjun Lu, and Wei Wang. 2005. Locating motifs in time-series data. In Advances in Knowledge Discovery and Data Mining, TuBao Ho, David Cheung, and Huan Liu (Eds.). Lecture Notes in Computer Science, Vol. 3518. Springer Berlin, 343--353. 11430919_41

Digital Library

[16]

Michael Lones. 2011. Sean luke: Essentials of metaheuristics. Genetic Program. Evol. Mach. 12, 3 (2011), 333--334.

Digital Library

[17]

David Minnen, Charles Isbell, Irfan Essa, and Thad Starner. 2007a. Detecting subdimensional motifs: An efficient algorithm for generalized multivariate pattern discovery. In Proceedings of the 7th IEEE International Conference on Data Mining (ICDM’07). IEEE Computer Society, Washington, DC, 601--606.

Digital Library

[18]

David Minnen, Charles L. Isbell, Irfan Essa, and Thad Starner. 2007b. Discovering multivariate motifs using subsequence density estimation and greedy mixture learning. In Proceedings of the 22nd National Conference on Artificial Intelligence - Volume 1 (AAAI’07). AAAI Press, 615--620. http://dl.acm.org/ citation.cfm?id=1619645.1619744

Digital Library

[19]

Yasser Mohammad and Toyoaki Nishida. 2014. Exact discovery of length-range motifs. In Intelligent Information and Database Systems, NgocThanh Nguyen, Boonwat Attachoo, Bogdan Trawiski, and Kulwadee Somboonviwat (Eds.). Lecture Notes in Computer Science, Vol. 8398. Springer International Publishing, 23--32.

[20]

Abdullah Mueen. 2015. Enumeration of time series motifs of all lengths. Knowledge and Information Systems Archive 45, 1 (Oct. 2015), 105--132.

Digital Library

[21]

Abdullah Mueen and Eamonn Keogh. 2010. Online discovery and maintenance of time series motifs. In Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD’10). ACM, New York, NY, 1089--1098.

Digital Library

[22]

Abdullah Mueen, Eamonn Keogh, and Nima Bigdely-Shamlo. 2009a. Finding time series motifs in disk-resident data. In Proceedings of the 9th IEEE International Conference on Data Mining (ICDM’09). IEEE Computer Society, Washington, DC, 367--376.

Digital Library

[23]

Abdullah Mueen, Eamonn J. Keogh, Qiang Zhu, Sydney Cash, and M. Brandon Westover. 2009b. Exact discovery of time series motifs. In Proceedings of the SIAM International Conference on Data Mining (SDM’09). SIAM, 12 Pages.

[24]

T. Oates. 2002. PERUSE: An unsupervised algorithm for finding recurring patterns in time series. In Proceedings of the IEEE International Conference on Data Mining, 2002 (ICDM’03). IEEE, 330--337.

Digital Library

[25]

Pranav Patel, Eamonn Keogh, Jessica Lin, and Stefano Lonardi. 2002. Mining motifs in massive time series databases. In Proceedings of the IEEE International Conference on Data Mining (ICDM’02). IEEE Computer Society, Washington, DC, 370--377. http://dl.acm.org/citation.cfm?id=844380.844710

Digital Library

[26]

Majed Sahli, Essam Mansour, and Panos Kalnis. 2014. ACME: A scalable parallel system for extracting frequent patterns from a very long sequence. VLDB J. 23, 6 (2014), 871--893. 10.1007/s00778-014-0370-1

[27]

Zeeshan Syed, Collin Stultz, Manolis Kellis, Piotr Indyk, and John Guttag. 2010. Motif discovery in physiological datasets: A methodology for inferring predictive elements. ACM Trans. Knowl. Discov. Data 4, 1, Article 2 (Jan. 2010), 23 pages.

Digital Library

[28]

S. Tata and J. M. Patel. 2008. FLAME: Shedding light on hidden frequent patterns in sequence datasets. In Proceedings of the 24th IEEE International Conference on Data Engineering (ICDE’08). IEEE, 1343--1345.

Digital Library

[29]

Alireza Vahdatpour, Navid Amini, and Majid Sarrafzadeh. 2009. Toward unsupervised activity discovery using multi-dimensional motif detection in time series. In Proceedings of the 21st International Joint Conference on Artifical Intelligence (IJCAI’09). Morgan Kaufmann Publishers Inc., San Francisco, CA, 1261--1266. http://dl.acm.org/citation.cfm?id=1661445.1661647

Digital Library

[30]

Dragomir Yankov, Eamonn Keogh, Jose Medina, Bill Chiu, and Victor Zordan. 2007. Detecting time series motifs under uniform scaling. In Proceedings of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD’07). ACM, New York, NY, 844--853.

Digital Library

[31]

MyatSu Yin, Songsri Tangsripairoj, and Benjarath Pupacdi. 2014. Variable length motif-based time series classification. In Recent Advances in Information and Communication Technology, Sirapat Boonkrong, Herwig Unger, and Phayung Meesad (Eds.). Advances in Intelligent Systems and Computing, Vol. 265. Springer International Publishing, 73--82.

[32]

S. Yingchareonthawornchai, H. Sivaraks, T. Rakthanmanon, and C. A. Ratanamahatana. 2013. Efficient proper length time series motif discovery. In Proceedings of the 13th IEEE International Conference on Data Mining (ICDM). IEEE, 1265--1270.

Cited By

Germain TTruong COudre L(2024)Persistence-Based Motif Discovery in Time SeriesIEEE Transactions on Knowledge and Data Engineering10.1109/TKDE.2024.341730336:11(6814-6827)Online publication date: 1-Nov-2024
https://dl.acm.org/doi/10.1109/TKDE.2024.3417303
Van Wesenbeeck DYurtman AMeert WBlockeel H(2024)LoCoMotif: discovering time-warped motifs in time seriesData Mining and Knowledge Discovery10.1007/s10618-024-01032-z38:4(2276-2305)Online publication date: 1-Jul-2024
https://dl.acm.org/doi/10.1007/s10618-024-01032-z
Al-Tahat MTrecakov SCook J(2022)AppEKG: A Simple Unifying View of HPC Applications in Production2022 IEEE/ACM International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS)10.1109/PMBS56514.2022.00017(129-134)Online publication date: Nov-2022
https://doi.org/10.1109/PMBS56514.2022.00017
Show More Cited By

Index Terms

Latent Time-Series Motifs

Recommendations

Probabilistic discovery of time series motifs
KDD '03: Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining

Several important time series data mining problems reduce to the core task of finding approximately repeated subsequences in a longer time series. In an earlier work, we formalized the idea of approximately repeated subsequences by introducing the ...
Finding functional promoter motifs by computational methods: a word of caution

The standard practice in the analysis of promoters is to select promoter regions of convenient length. This may lead to false results when searching for Transcription Factor Binding Sites (TFBSs), since the sequences may contain coding segments. In such ...
An Efficient Method for Discovering Variable-length Motifs in Time Series based on Suffix Array
SoICT '19: Proceedings of the 10th International Symposium on Information and Communication Technology

Repeated patterns in time series, also called motifs, are approximately repeated subsequences embedded within a longer time series data. There have been some popular and effective algorithms for discovering motifs in time series. Nevertheless, these ...

Comments

Information & Contributors

Information

Published In

cover image ACM Transactions on Knowledge Discovery from Data

ACM Transactions on Knowledge Discovery from Data Volume 11, Issue 1

February 2017

288 pages

ISSN:1556-4681

EISSN:1556-472X

DOI:10.1145/2974720

Editor:
Philip S. Yu
University of Illinois at Chicago, USA

Issue’s Table of Contents

Copyright © 2016 ACM.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 20 July 2016

Accepted: 01 May 2016

Revised: 01 April 2016

Received: 01 June 2015

Published in TKDD Volume 11, Issue 1

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article
Research
Refereed

Funding Sources

Seventh Framework Programme of the European Commission, through the REDUCTION
Deutsche Forschungsgemeinschaft within the project HyLAP

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

35
Total Citations
View Citations
681
Total Downloads

Downloads (Last 12 months)25
Downloads (Last 6 weeks)1

Reflects downloads up to 08 Mar 2025

Other Metrics

View Author Metrics

Citations

Cited By

Germain TTruong COudre L(2024)Persistence-Based Motif Discovery in Time SeriesIEEE Transactions on Knowledge and Data Engineering10.1109/TKDE.2024.341730336:11(6814-6827)Online publication date: 1-Nov-2024
https://dl.acm.org/doi/10.1109/TKDE.2024.3417303
Van Wesenbeeck DYurtman AMeert WBlockeel H(2024)LoCoMotif: discovering time-warped motifs in time seriesData Mining and Knowledge Discovery10.1007/s10618-024-01032-z38:4(2276-2305)Online publication date: 1-Jul-2024
https://dl.acm.org/doi/10.1007/s10618-024-01032-z
Al-Tahat MTrecakov SCook J(2022)AppEKG: A Simple Unifying View of HPC Applications in Production2022 IEEE/ACM International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS)10.1109/PMBS56514.2022.00017(129-134)Online publication date: Nov-2022
https://doi.org/10.1109/PMBS56514.2022.00017
Ermshaus ASchäfer PLeser U(2022)Window Size Selection in Unsupervised Time Series Analytics: A Review and BenchmarkAdvanced Analytics and Learning on Temporal Data10.1007/978-3-031-24378-3_6(83-101)Online publication date: 19-Sep-2022
https://dl.acm.org/doi/10.1007/978-3-031-24378-3_6
Gao JYing XXu CWang JZhang SLi Z(2021)Graph-Based Stock Recommendation by Time-Aware Relational Attention NetworkACM Transactions on Knowledge Discovery from Data10.1145/345139716:1(1-21)Online publication date: 20-Jul-2021
https://dl.acm.org/doi/10.1145/3451397
Peng BFatourou PPalpanas T(2021)Fast data series indexing for in-memory dataThe VLDB Journal — The International Journal on Very Large Data Bases10.1007/s00778-021-00677-230:6(1041-1067)Online publication date: 18-Jun-2021
https://dl.acm.org/doi/10.1007/s00778-021-00677-2
Boniol PLinardi MRoncallo FPalpanas TMeftah MRemy E(2021)Unsupervised and scalable subsequence anomaly detection in large data seriesThe VLDB Journal — The International Journal on Very Large Data Bases10.1007/s00778-021-00655-830:6(909-931)Online publication date: 3-Mar-2021
https://dl.acm.org/doi/10.1007/s00778-021-00655-8
Linardi MZhu YPalpanas TKeogh E(2020)Matrix profile goes MAD: variable-length motif and discord discovery in data seriesData Mining and Knowledge Discovery10.1007/s10618-020-00685-w34:4(1022-1071)Online publication date: 7-May-2020
https://doi.org/10.1007/s10618-020-00685-w
Feng TNarayanan S(2019)Discovering Optimal Variable-length Time Series Motifs in Large-scale Wearable Recordings of Human Bio-behavioral SignalsICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)10.1109/ICASSP.2019.8682427(7615-7619)Online publication date: May-2019
https://doi.org/10.1109/ICASSP.2019.8682427
Hashida STamura K(2019)MACD-Histogram-based Fully Convolutional Neural Networks for Classifying Time Series2019 6th International Conference on Control, Decision and Information Technologies (CoDIT)10.1109/CoDIT.2019.8820629(1049-1054)Online publication date: Apr-2019
https://doi.org/10.1109/CoDIT.2019.8820629
Show More Cited By

View Options

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Article

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Figures

Tables

Media

View Issue’s Table of Contents