Advertisement

Unsupervised online change point detection in high-dimensional time series

  • Masoomeh ZameniEmail author
  • Amin Sadri
  • Zahra Ghafoori
  • Masud Moshtaghi
  • Flora D. Salim
  • Christopher Leckie
  • Kotagiri Ramamohanarao
Regular Paper
  • 76 Downloads

Abstract

A critical problem in time series analysis is change point detection, which identifies the times when the underlying distribution of a time series abruptly changes. However, several shortcomings limit the use of some existing techniques in real-world applications. First, several change point detection techniques are offline methods, where the whole time series needs to be stored before change point detection can be performed. These methods are not applicable to streaming time series. Second, most techniques assume that the time series is low-dimensional and hence have problems handling high-dimensional time series, where not all dimensions may cause the change. Finally, most methods require user-defined parameters that need to be chosen based on the observed data, which limits their applicability to new unseen data. To address these issues, we propose an Information Gain-based method that does not require prior distributional knowledge for detecting change points and handles high-dimensional time series. The advantages of our proposed method compared to the state-of-the-art algorithms are demonstrated from theoretical basis, as well as via experiments on four synthetic and three real-world human activity datasets.

Keywords

Time series Online change point detection Segmentation Information Gain Theory 

Notes

References

  1. 1.
    Aminikhanghahi S, Cook DJ (2016) A survey of methods for time series change point detection. Knowl Inf Syst 2(3):1–29Google Scholar
  2. 2.
    Aminikhanghahi S, Cook DJ (2017) Using change point detection to automate daily activity segmentation. In: IEEE international conference on pervasive computing and communications workshops (PerCom Workshops), pp 262–267Google Scholar
  3. 3.
    Appel U, Brandt AV (1983) Adaptive sequential segmentation of piecewise stationary time series. Inf Sci 29(1):27–56CrossRefzbMATHGoogle Scholar
  4. 4.
    Badarna M, Wolff R (2014) Fast and accurate detection of changes in data streams. Stat Anal Data Min ASA Data Sci J 7(2):125–139MathSciNetCrossRefGoogle Scholar
  5. 5.
    Barddal JP, Gomes HM, Enembreck F, Pfahringer B, Bifet A (2016) On dynamic feature weighting for feature drifting data streams. In: Joint european conference on machine learning and knowledge discovery in databases, pp 129–144Google Scholar
  6. 6.
    Barddal JP, Gomes HM, Enembreck F, Pfahringer B (2017) A survey on feature drift adaptation: definition, benchmark, challenges and future directions. J Syst Softw 127:278–294CrossRefGoogle Scholar
  7. 7.
    Barddal JP, Enembreck F, Gomes HM, Bifet A, Pfahringer B (2019a) Boosting decision stumps for dynamic feature selection on data streams. Inf Syst 83:13–29CrossRefGoogle Scholar
  8. 8.
    Barddal JP, Enembreck F, Gomes HM, Bifet A, Pfahringer B (2019b) Merit-guided dynamic feature selection filter for data streams. Expert Syst Appl 116:227–242CrossRefGoogle Scholar
  9. 9.
    Blythe DA, Von Bunau P, Meinecke FC, Muller K-R (2012) Feature extraction for change-point detection using stationary subspace analysis. IEEE Trans Neural Netw Learn Syst 23(4):631–643CrossRefGoogle Scholar
  10. 10.
    Chakraborty K, Mehrotra K, Mohan CK, Ranka S (1992) Forecasting the behavior of multivariate time series using neural networks. Neural Netw 5(6):961–970CrossRefGoogle Scholar
  11. 11.
    Dasu T, Krishnan S, Venkatasubramanian S, Yi K (2006) An information-theoretic approach to detecting changes in multi-dimensional data streams. In: Proceedings of the 38th symposium on interface of statistics, computing science, and applications (Interface ’06), PasadenaGoogle Scholar
  12. 12.
    Desobry F, Davy M, Doncarli C (2005) An online kernel change detection algorithm. IEEE Trans Signal Process 53(8):2961–2974MathSciNetCrossRefzbMATHGoogle Scholar
  13. 13.
    Feuz KD, Cook DJ, Rosasco C, Robertson K, Schmitter-Edgecombe M (2015) Automated detection of activity transitions for prompting. IEEE Trans Hum Mach Syst 45(5):575–585CrossRefGoogle Scholar
  14. 14.
    Fryzlewicz P (2014) Wild binary segmentation for multiple change-point detection. Ann Stat 42(6):2243–2281MathSciNetCrossRefzbMATHGoogle Scholar
  15. 15.
    Gharghabi S, Ding Y, Yeh C-CM, Kamgar K, Ulanova L, Keogh E (2017) Matrix profile VIII: domain agnostic online semantic segmentation at superhuman performance levels. In: IEEE international conference on data mining (ICDM), pp 117–126Google Scholar
  16. 16.
    Gharghabi S, Yeh C-CM, Ding Y, Ding W, Hibbing P, LaMunion S, Kaplan A, Crouter SE, Keogh E (2018) Domain agnostic online semantic segmentation for multi-dimensional time series. In: Data mining and knowledge discovery, pp 1–35Google Scholar
  17. 17.
    Hotelling H (1931) The generalization of student’s ratio. Ann Math Stat 2(3):360–378CrossRefzbMATHGoogle Scholar
  18. 18.
    Kawahara Y, Sugiyama M (2012) Sequential change-point detection based on direct density-ratio estimation. Stat Anal Data Min 5(2):114–127MathSciNetCrossRefGoogle Scholar
  19. 19.
    Kifer D, Ben-David S, Gehrke J (2004) Detecting change in data streams. In: Proceedings of the thirtieth international conference on very large data bases, vol 30, pp 180–191Google Scholar
  20. 20.
    Korkas KK, Fryzlewicz P (2017) Multiple change-point detection for non-stationary time series using wild binary segmentation. Stat Sin 27(1):287–311MathSciNetzbMATHGoogle Scholar
  21. 21.
    Krishnan NC, Cook DJ (2014) Activity recognition on streaming sensor data. Pervasive Mobile Comput 10:138–154CrossRefGoogle Scholar
  22. 22.
    Kuncheva LI (2013) Change detection in streaming multivariate data using likelihood detectors. IEEE Trans Knowl Data Eng 25(5):1175–1180CrossRefGoogle Scholar
  23. 23.
    Kurt MN, Raghavan V, Wang X (2017) Multi-sensor sequential change detection with unknown change propagation dynamics. arXiv preprint arXiv:1708.04722
  24. 24.
    Liu S, Yamada M, Collier N, Sugiyama M (2013) Change-point detection in time-series data by relative density-ratio estimation. Neural Netw 43:72–83CrossRefzbMATHGoogle Scholar
  25. 25.
    Matteson DS, James NA (2014) A nonparametric approach for multiple change point analysis of multivariate data. Am Stat Assoc 109(505):334–345MathSciNetCrossRefzbMATHGoogle Scholar
  26. 26.
    Moshtaghi M, Erfani S, Leckie C, Bezdek J (2017) Exponentially weighted ellipsoidal model for anomaly detection. Int J Intell Syst 32(9):881–899CrossRefGoogle Scholar
  27. 27.
    Nason GP, Von Sachs R, Kroisandt G (2000) Wavelet processes and adaptive estimation of the evolutionary wavelet spectrum. R Stat Soc Ser B (Stat Methodol) 62(2):271–292MathSciNetCrossRefGoogle Scholar
  28. 28.
    Nguyen H-L, Woon Y-K, Ng W-K, Wan L (2012) Heterogeneous ensemble for feature drifts in data streams. In: Pacific-Asia conference on knowledge discovery and data mining, pp 1–12Google Scholar
  29. 29.
    Noor MHM, Salcic Z, Kevin I, Wang K (2017) Adaptive sliding window segmentation for physical activity recognition using a single tri-axial accelerometer. Pervasive Mobile Comput 38:41–59CrossRefGoogle Scholar
  30. 30.
    Ordóñez FJ, Roggen D (2016) Deep convolutional and lstm recurrent neural networks for multimodal wearable activity recognition. Sensors 16(1):115CrossRefGoogle Scholar
  31. 31.
    Rajasegarar S, Bezdek JC, Moshtaghi M, Leckie C, Havens TC, Palaniswami M (2012) Measures for clustering and anomaly detection in sets of higher dimensional ellipsoids. In: IEEE international joint conference on neural networks (IJCNN), pp 1–8Google Scholar
  32. 32.
    Sadri A, Ren Y, Salim FD (2017) Information gain-based metric for recognizing transitions in human activities. Pervasive Mobile Comput 38:92–109CrossRefGoogle Scholar
  33. 33.
  34. 34.
    San-Segundo R, Lorenzo-Trueba J, Martínez-González B, Pardo JM (2016) Segmenting human activities based on HMMs using smartphone inertial sensors. Pervasive Mobile Comput 30:84–96CrossRefGoogle Scholar
  35. 35.
    Siegmund D, Venkatraman E (1995) Using the generalized likelihood ratio statistic for sequential detection of a change-point. Ann Stat 23:255–271MathSciNetCrossRefzbMATHGoogle Scholar
  36. 36.
    Vana P (2015) Blind segmentation of time-series: a two-level approach. PhD thesis, TU Delft, Delft University of TechnologyGoogle Scholar
  37. 37.
    Yahmed YB, Bakar AA, Hamdan AR, Ahmed A, Abdullah SMS (2015) Adaptive sliding window algorithm for weather data segmentation. Theor Appl Inf Technol 80(2):322Google Scholar
  38. 38.
    Yamada M, Kimura A, Naya F, Sawada H (2013) Change-point detection with feature selection in high-dimensional time-series data. In: international joint conference on artificial intelligence (IJCAI), pp 1827–1833Google Scholar
  39. 39.
    Yao L, Sheng QZ, Ruan W, Li X, Wang S, Yang Z (2015) Unobtrusive posture recognition via online learning of multi-dimensional rfid received signal strength. In: IEEE 21st international conference on parallel and distributed systems (ICPADS), pp 116–123Google Scholar
  40. 40.
    Yeh C-CM, Zhu Y, Ulanova L, Begum N, Ding Y, Dau HA, Silva DF, Mueen A, Keogh E (2016) Matrix profile I: all pairs similarity joins for time series: a unifying view that includes motifs, discords and shapelets. In: IEEE 16th international conference on data mining (ICDM), pp 1317–1322Google Scholar
  41. 41.
    Zhang M, Sawchuk AA (2012) Usc-had: a daily activity dataset for ubiquitous activity recognition using wearable sensors. In: ACM international conference on ubiquitous computing (UbiComp) workshop on situation, activity and goal awareness (SAGAware). Pittsburgh, Pennsylvania, USA, pp 1036–1043Google Scholar

Copyright information

© Springer-Verlag London Ltd., part of Springer Nature 2019

Authors and Affiliations

  1. 1.School of Computing and Information SystemsUniversity of MelbourneMelbourneAustralia
  2. 2.School of ScienceRMIT UniversityMelbourneAustralia

Personalised recommendations