Filters








755 Hits in 5.4 sec

A disk-aware algorithm for time series motif discovery

Abdullah Mueen, Eamonn Keogh, Qiang Zhu, Sydney S. Cash, M. Brandon Westover, Nima Bigdely-Shamlo
2010 Data mining and knowledge discovery  
In this work, we leverage off previous work on pivot-based indexing to introduce a disk-aware algorithm to find time series motifs exactly in multi-gigabyte databases which contain on the order of tens  ...  Time series motifs are sets of very similar subsequences of a long time series.  ...  In this work, we describe for the first time a disk-aware algorithm to find time series motifs in multi-gigabyte databases containing tens of millions of As we shall show, our algorithm allows us to tackle  ... 
doi:10.1007/s10618-010-0176-8 pmid:32153346 pmcid:PMC7062370 fatcat:i3qgkvu4mfavpeyvdyvmzkfe3y

Finding Time Series Motifs in Disk-Resident Data

Abdullah Mueen, Eamonn Keogh, Nima Bigdely-Shamlo
2009 2009 Ninth IEEE International Conference on Data Mining  
In this work, we describe for the first time a disk-aware algorithm to find exact time series motifs in multi-gigabyte databases which contain on the order of tens of millions of time series.  ...  Time series motifs are sets of very similar subsequences of a long time series.  ...  In this work, we describe for the first time a disk-aware algorithm to find exact time series motifs in multi-gigabyte databases containing tens of millions of time series.  ... 
doi:10.1109/icdm.2009.15 dblp:conf/icdm/MueenKS09 fatcat:cw4dssycjfar7gh7wqmx22gvou

Multiresolution Motif Discovery in Time Series [chapter]

Nuno Castro, Paulo Azevedo
2010 Proceedings of the 2010 SIAM International Conference on Data Mining  
Time series motif discovery is an important problem with applications in a variety of areas that range from telecommunications to medicine. Several algorithms have been proposed to solve the problem.  ...  Our approach is scalable and disk-efficient since it only needs one single pass over the time series database.  ...  The exact motif discovery solution for a time series presents quadratic complexity.  ... 
doi:10.1137/1.9781611972801.73 dblp:conf/sdm/CastroA10 fatcat:3rvvm3kzcrc6bc5alzbzzdzdku

Effective and Efficient Variable-Length Data Series Analytics [article]

Michele Linardi
2020 arXiv   pre-print
In this Ph.D. work, we present the first solutions that inherently support scalable and variable-length similarity search in data series, applied to sequence/subsequences matching, motif and discord discovery  ...  Unfortunately, the obvious brute-force solution, which provides an outcome for all lengths within a given range is computationally untenable.  ...  The first one, DAD (Disk aware discord discovery) [27] , implements an algorithm suitable to enumerate the fixed-length T op − 1 m th discords.  ... 
arXiv:2009.11648v1 fatcat:yh56dcspevcwxjoi4ifixginoa

Exact Primitives for Time Series Data Mining

Sbdullah Al Mueen
2012 Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '12  
In this work, we describe for the first time a disk-aware algorithm to find exact time series motifs in multi-gigabyte databases containing tens of millions of time series.  ...  DAME: Disk Aware Motif Enumeration A set of time series of length n can be thought of as a set of points in n-dimensional space.  ...  for l ← 1 to m do every possible length 6: for i ← 1 to |S| − l + 1 do every start position 7: for k ← 1 to |D| do compute distances of every time series to the candidate shapelet S i,l 8: sort(L)  ... 
doi:10.1145/2339530.2378374 fatcat:epmxgzahtfglnmfvyuqc4qlt6u

Modeling Content and Users: Structured Probabilistic Representation and Scalable Inference Algorithms

Amr Ahmed
2012 Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '12  
In this work, we describe for the first time a disk-aware algorithm to find exact time series motifs in multi-gigabyte databases containing tens of millions of time series.  ...  DAME: Disk Aware Motif Enumeration A set of time series of length n can be thought of as a set of points in n-dimensional space.  ...  for l ← 1 to m do every possible length 6: for i ← 1 to |S| − l + 1 do every start position 7: for k ← 1 to |D| do compute distances of every time series to the candidate shapelet S i,l 8: sort(L)  ... 
doi:10.1145/2339530.2378373 fatcat:2tbhm22bujd5lcqx752epdb27e

Online Discovery of Top-k Similar Motifs in Time Series Data [chapter]

Hoang Thanh Lam, Ninh Dang Pham, Toon Calders
2011 Proceedings of the 2011 SIAM International Conference on Data Mining  
A motif is a pair of non-overlapping sequences with very similar shapes in a time series. We study the online topk most similar motif discovery problem.  ...  We also show possible application of the top-k similar motifs discovery problem.  ...  Acknowledgements We deeply thank A. Mueen and professor E. Keogh for their released datasets, source code and useful discussion in the early stage of the project.  ... 
doi:10.1137/1.9781611972818.86 dblp:conf/sdm/LamCP11 fatcat:5xxoszjqfndjxfvmov6deqvu4u

Matrix Profile II: Exploiting a Novel Algorithm and GPUs to Break the One Hundred Million Barrier for Time Series Motifs and Joins

Yan Zhu, Zachary Zimmerman, Nader Shakibay Senobari, Chin-Chia Michael Yeh, Gareth Funning, Abdullah Mueen, Philip Brisk, Eamonn Keogh
2016 2016 IEEE 16th International Conference on Data Mining (ICDM)  
In this work we show that a combination of a novel algorithm and a high-performance GPU allows us to significantly improve the scalability of motif discovery.  ...  We demonstrate the scalability of our ideas by finding the full set of exact motifs on a dataset with one hundred million subsequences, by far the largest dataset ever mined for time series motifs.  ...  Motif Discovery Background Motif discovery for time series was introduced in 2003 [5] (although the classic paper of Agrawal, Faloutsos and Swami foreshadows motifs by computing all-pair similarity for  ... 
doi:10.1109/icdm.2016.0085 dblp:conf/icdm/ZhuZSYFMBK16 fatcat:am6llbgs4nghnlsi3s3jsyp344

Time series joins, motifs, discords and shapelets: a unifying view that exploits the matrix profile

Chin-Chia Michael Yeh, Yan Zhu, Liudmila Ulanova, Nurjahan Begum, Yifei Ding, Hoang Anh Dau, Zachary Zimmerman, Diego Furtado Silva, Abdullah Mueen, Eamonn Keogh
2017 Data mining and knowledge discovery  
We demonstrate the utility of our ideas for many time series data mining problems, including motif discovery, novelty discovery, shapelet discovery, semantic segmentation, density estimation, and contrast  ...  However, there has been incidentally provides the fastest known algorithm for both these extensively-studied problems.  ...  It is clear that a scalable TSAPSS algorithm would be a versatile building block for developing algorithms for many time series data mining tasks (e.g., motif discovery, shapelet discovery, semantic segmentation  ... 
doi:10.1007/s10618-017-0519-9 fatcat:r6p5lk6td5dvfltcj2vlucnf3q

Matrix Profile Goes MAD: Variable-Length Motif And Discord Discovery in Data Series [article]

Michele Linardi and Yan Zhu and Themis Palpanas and Eamonn Keogh
2020 arXiv   pre-print
In this work, we introduce a new framework, which provides an exact and scalable motif and discord discovery algorithm that efficiently finds all motifs and discords in a given range of lengths.  ...  In the last fifteen years, data series motif and discord discovery have emerged as two useful and well-used primitives for data series mining, with applications to many domains, including robotics, entomology  ...  The first one, DAD (Disk Aware Discord Discovery) (Yankov et al., 2007b) , implements an algorithm suitable for enumerating the fixed-length m th discords of a data series collection stored on a disk.  ... 
arXiv:2008.13447v1 fatcat:qpfredpj6vh4jljuyigbzq6q3y

Exact Discovery of Time Series Motifs [chapter]

Abdullah Mueen, Eamonn Keogh, Qiang Zhu, Sydney Cash, Brandon Westover
2009 Proceedings of the 2009 SIAM International Conference on Data Mining  
In this work, for the first time, we show a tractable exact algorithm to find time series motifs.  ...  Time series motifs are pairs of individual time series, or subsequences of a longer time series, which are very similar to each other.  ...  force motif discovery Algorithm Brute Force Motif Discovery Procedure [L 1 ,L 2 ]=BruteForce_Motif (D) in: D: Database of Time Series Out: L 1 ,L 2 : Locations for a Motif 1 best-so-far = INF  ... 
doi:10.1137/1.9781611972795.41 pmid:31656693 pmcid:PMC6814436 dblp:conf/sdm/MueenKZCW09 fatcat:vif6qlq2m5c6bebz7mm5xssxhm

Exploiting a novel algorithm and GPUs to break the ten quadrillion pairwise comparisons barrier for time series motifs and joins

Yan Zhu, Zachary Zimmerman, Nader Shakibay Senobari, Chin-Chia Michael Yeh, Gareth Funning, Abdullah Mueen, Philip Brisk, Eamonn Keogh
2017 Knowledge and Information Systems  
Time series motifs are approximately repeated subsequences found within a longer time series.  ...  Zhu et al. this work, we demonstrate that a combination of a novel algorithm and a high-performance GPU allows us to significantly improve the scalability of motif discovery.  ...  He is especially interested in scalable algorithms for real-time data.  ... 
doi:10.1007/s10115-017-1138-x fatcat:p6zrlvn5wnfipk6yhi76pir6pu

Discovering Multidimensional Motifs in Physiological Signals for Personalized Healthcare

Arvind Balasubramanian, Jun Wang, Balakrishnan Prabhakaran
2016 IEEE Journal on Selected Topics in Signal Processing  
Performance evaluation using synthetic and real body sensor data in unsupervised motif discovery tasks shows that the approach is effective for (a) concurrent processing of multidimensional time series  ...  In this paper, we propose an efficient real-time approach to MDM discovery in body sensor generated time series data for monitoring performance of patients during therapy.  ...  Unidimensional Motif Discovery For motif discovery in individual dimensions of the multidimensional time series, any stateof-the-art motif discovery technique could be used.  ... 
doi:10.1109/jstsp.2016.2543679 pmid:28191269 pmcid:PMC5298205 fatcat:ux3mrjpevffohka7qowrrw5i7i

A symbolic representation of time series, with implications for streaming algorithms

Jessica Lin, Eamonn Keogh, Stefano Lonardi, Bill Chiu
2003 Proceedings of the 8th ACM SIGMOD workshop on Research issues in data mining and knowledge discovery - DMKD '03  
This last feature explicitly thwarts efforts to use the representations with streaming algorithms. In this work we introduce a new symbolic representation of time series.  ...  The main reason for this apparent paradox is the fact that the vast majority of work on streaming data explicitly assumes that the data is discrete, whereas the vast majority of time series data is real  ...  tasks such as anomaly detection and motif discovery.  ... 
doi:10.1145/882085.882086 fatcat:u5tlfeazdnd4loki25hvpxkn2e

A symbolic representation of time series, with implications for streaming algorithms

Jessica Lin, Eamonn Keogh, Stefano Lonardi, Bill Chiu
2003 Proceedings of the 8th ACM SIGMOD workshop on Research issues in data mining and knowledge discovery - DMKD '03  
This last feature explicitly thwarts efforts to use the representations with streaming algorithms. In this work we introduce a new symbolic representation of time series.  ...  The main reason for this apparent paradox is the fact that the vast majority of work on streaming data explicitly assumes that the data is discrete, whereas the vast majority of time series data is real  ...  tasks such as anomaly detection and motif discovery.  ... 
doi:10.1145/882082.882086 dblp:conf/dmkd/LinKLC03 fatcat:t6q6bery7vc6fl3h4dc2w5a2qy
« Previous Showing results 1 — 15 out of 755 results