k-Shape: Efficient and Accurate Clustering of Time Series

The proliferation and ubiquity of temporal data across many disciplines has generated substantial interest in the analysis and mining of time series. Clustering is one of the most popular data mining methods, not only due to its exploratory power, but also as a preprocessing step or subroutine for other techniques. In this paper, we describe k-Shape, a novel algorithm for time-series clustering. k-Shape relies on a scalable iterative refinement procedure, which creates homogeneous and well-separated clusters. As its distance measure, k-Shape uses a normalized version of the cross-correlation measure in order to consider the shapes of time series while comparing them. Based on the properties of that distance measure, we develop a method to compute cluster centroids, which are used in every iteration to update the assignment of time series to clusters. An extensive experimental evaluation against partitional, hierarchical, and spectral clustering methods, with the most competitive distance measures, showed the robustness of k-Shape. Overall, k-Shape emerges as a domain-independent, highly accurate, and efficient clustering approach for time series with broad applications.

[1]  Eamonn J. Keogh,et al.  Logical-shapelets: an expressive primitive for time series classification , 2011, KDD.

[2]  Francesco Masulli,et al.  A survey of kernel and spectral methods for clustering , 2008, Pattern Recognit..

[3]  Eamonn J. Keogh,et al.  Time Series Epenthesis: Clustering Time Series Streams Requires Ignoring Some Data , 2011, 2011 IEEE 11th International Conference on Data Mining.

[4]  Michalis Vazirgiannis,et al.  c ○ 2001 Kluwer Academic Publishers. Manufactured in The Netherlands. On Clustering Validation Techniques , 2022 .

[5]  Eamonn J. Keogh,et al.  Making Time-Series Classification More Accurate Using Learned Constraints , 2004, SDM.

[6]  T. Warren Liao,et al.  Clustering of time series data - a survey , 2005, Pattern Recognit..

[7]  L. Gupta,et al.  Nonlinear alignment and averaging for estimating the evoked potential , 1996, IEEE Transactions on Biomedical Engineering.

[8]  M. Friedman.  The Use of Ranks to Avoid the Assumption of Normality Implicit in the Analysis of Variance , 1937 .

[9]  Hao Wang,et al.  Durable Queries over Historical Time Series , 2014, IEEE Transactions on Knowledge and Data Engineering.

[10]  Eamonn J. Keogh,et al.  Time series shapelets: a new primitive for data mining , 2009, KDD.

[11]  Dina Q. Goldin,et al.  On Similarity Queries for Time-Series Data: Constraint Specification and Implementation , 1995, CP.

[12]  Aristides Gionis,et al.  Correlating financial time series with micro-blogging activity , 2012, WSDM '12.

[13]  Eamonn J. Keogh,et al.  Searching and Mining Trillions of Time Series Subsequences under Dynamic Time Warping , 2012, KDD.

[14]  Ali S. Hadi,et al.  Finding Groups in Data: An Introduction to Chster Analysis , 1991 .

[15]  Eamonn J. Keogh,et al.  CID: an efficient complexity-invariant distance for time series , 2013, Data Mining and Knowledge Discovery.

[16]  Shuai Wang,et al.  Mining of Moving Objects from Time-Series Images and its Application to Satellite Weather Imagery , 2004, Journal of Intelligent Information Systems.

[17]  Steve Goddard,et al.  Geospatial decision support for drought risk management , 2003, CACM.

[18]  Gareth J. Janacek,et al.  Clustering time series from ARMA models with clipped data , 2004, KDD.

[19]  S. Chiba,et al.  Dynamic programming algorithm optimization for spoken word recognition , 1978 .

[20]  Ge Yu,et al.  Similarity Match Over High Speed Time-Series Streams , 2007, 2007 IEEE 23rd International Conference on Data Engineering.

[21]  Donald J. Berndt,et al.  Using Dynamic Time Warping to Find Patterns in Time Series , 1994, KDD Workshop.

[22]  Dimitrios Gunopulos,et al.  Indexing Multidimensional Time-Series , 2004, The VLDB Journal.

[23]  Dimitrios Gunopulos,et al.  Discovering similar multidimensional trajectories , 2002, Proceedings 18th International Conference on Data Engineering.

[24]  Jignesh M. Patel,et al.  An efficient and accurate method for evaluating time series similarity , 2007, SIGMOD '07.

[25]  Konstantinos Kalpakis,et al.  Distance measures for effective clustering of ARIMA time-series , 2001, Proceedings 2001 IEEE International Conference on Data Mining.

[26]  Hui Ding,et al.  Querying and mining of time series data: experimental comparison of representations and distance measures , 2008, Proc. VLDB Endow..

[27]  Gene H. Golub,et al.  Matrix computations , 1983 .

[28]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .

[29]  Yunhao Liu,et al.  Indexable PLA for Efficient Similarity Search , 2007, VLDB.

[30]  Nikos Mamoulis,et al.  Fast and Exact Warping of Time Series Using Adaptive Segmental Approximations , 2005, Machine Learning.

[31]  Jure Leskovec,et al.  Patterns of temporal variation in online media , 2011, WSDM '11.

[32]  Philip S. Yu,et al.  MALM: a framework for mining sequence database at multiple abstraction levels , 1998, CIKM '98.

[33]  Tim Oates,et al.  Identifying distinctive subsequences in multivariate time series by clustering , 1999, KDD '99.

[34]  Heikki Mannila,et al.  Rule Discovery from Time Series , 1998, KDD.

[35]  Yannis Theodoridis,et al.  Index-based Most Similar Trajectory Search , 2007, 2007 IEEE 23rd International Conference on Data Engineering.

[36]  Pierre Gançarski,et al.  A global averaging method for dynamic time warping, with applications to clustering , 2011, Pattern Recognit..

[37]  Qiang Wang,et al.  A multiresolution symbolic representation of time series , 2005, 21st International Conference on Data Engineering (ICDE'05).

[38]  R. Mantegna.  Hierarchical structure in financial markets , 1998, cond-mat/9802256.

[39]  Eamonn J. Keogh,et al.  Experimental comparison of representation methods and distance measures for time series data , 2010, Data Mining and Knowledge Discovery.

[40]  Dit-Yan Yeung,et al.  Mixtures of ARMA models for model-based time series clustering , 2002, 2002 IEEE International Conference on Data Mining, 2002. Proceedings..

[41]  Lei Chen,et al.  Robust and fast similarity search for moving object trajectories , 2005, SIGMOD '05.

[42]  Christos Faloutsos,et al.  Efficient Similarity Search In Sequence Databases , 1993, FODO.

[43]  Kuniaki Uehara,et al.  Extraction of Primitive Motion and Discovery of Association Rules from Human Motion Data , 2002, Progress in Discovery Science.

[44]  Dimitrios Gunopulos,et al.  Embedding-based subsequence matching in time-series databases , 2011, TODS.

[45]  Raymond T. Ng,et al.  Indexing spatio-temporal trajectories with Chebyshev polynomials , 2004, SIGMOD '04.

[46]  Vit Niennattrakul,et al.  Shape-Based Clustering for Time Series Data , 2012, PAKDD.

[47]  R. E. Lee,et al.  Distribution-free multiple comparisons between successive treatments , 1995 .

[48]  Lei Chen,et al.  On The Marriage of Lp-norms and Edit Distance , 2004, VLDB.

[49]  Eamonn J. Keogh,et al.  Time Series Classification under More Realistic Assumptions , 2013, SDM.

[50]  Eamonn J. Keogh,et al.  Clustering Time Series Using Unsupervised-Shapelets , 2012, 2012 IEEE 12th International Conference on Data Mining.

[51]  Steven G. Johnson,et al.  The Design and Implementation of FFTW3 , 2005, Proceedings of the IEEE.

[52]  Y. Katznelson.  An Introduction to Harmonic Analysis: Interpolation of Linear Operators , 1968 .

[53]  Anthony K. H. Tung,et al.  SpADe: On Shape-based Pattern Detection in Streaming Time Series , 2007, 2007 IEEE 23rd International Conference on Data Engineering.

[54]  R. Suganya,et al.  Data Mining Concepts and Techniques , 2010 .

[55]  Eamonn J. Keogh.  A decade of progress in indexing and mining large time series databases , 2006, VLDB.

[56]  Eamonn J. Keogh,et al.  Locally adaptive dimensionality reduction for indexing large time series databases , 2001, SIGMOD '01.

[57]  Gustavo E. A. P. A. Batista,et al.  An Empirical Comparison of Dissimilarity Measures for Time Series Classification , 2013, 2013 Brazilian Conference on Intelligent Systems.

[58]  Dragomir Anguelov,et al.  Mining The Stock Market : Which Measure Is Best ? , 2000 .

[59]  Tommi S. Jaakkola,et al.  A new approach to analyzing gene expression time series data , 2002, RECOMB '02.

[60]  Vladimir Pavlovic,et al.  Discovering clusters in motion time-series data , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[61]  Ada Wai-Chee Fu,et al.  Efficient time series matching by wavelets , 1999, Proceedings 15th International Conference on Data Engineering (Cat. No.99CB36337).

[62]  Michael I. Jordan,et al.  On Spectral Clustering: Analysis and an algorithm , 2001, NIPS.

[63]  Eamonn J. Keogh,et al.  Exact indexing of dynamic time warping , 2002, Knowledge and Information Systems.

[64]  Luis Gravano,et al.  k-Shape: Efficient and Accurate Clustering of Time Series , 2015, SIGMOD Conference.

[65]  William M. Rand,et al.  Objective Criteria for the Evaluation of Clustering Methods , 1971 .

[66]  R. Stephenson.  A and V , 1962, The British journal of ophthalmology.

[67]  Eamonn J. Keogh,et al.  Clustering of time-series subsequences is meaningless: implications for previous and future research , 2004, Knowledge and Information Systems.

[68]  Clu-istos Foutsos,et al.  Fast subsequence matching in time-series databases , 1994, SIGMOD '94.

[69]  Christos Faloutsos,et al.  Efficiently supporting ad hoc queries in large datasets of time sequences , 1997, SIGMOD '97.

[70]  J. Tukey,et al.  An algorithm for the machine calculation of complex Fourier series , 1965 .

[71]  C. Ratanamahatana,et al.  Shape averaging under Time Warping , 2009, 2009 6th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology.