ABSTRACT
Dynamic Time Warping (DTW) is a widely used distance measure for time series data mining. Its quadratic complexity requires the application of various techniques (e.g. warping constraints, lower-bounds) for deployment in real-time scenarios. In this paper we propose a randomized greedy warping algorithm for finding similarity between time series instances. We show that the proposed algorithm outperforms the simple greedy approach and also provides very good time series similarity approximation consistently, as compared to DTW. We show that the Randomized Time Warping (RTW) can be used in place of DTW as a fast similarity approximation technique by trading some classification accuracy for very fast classification.
- C. C. Aggarwal, A. Hinneburg, and D. A. Keim. On the surprising behavior of distance metrics in high dimensional space. In Database Theory - ICDT 2001, pages 420--434. Springer Berlin Heidelberg, 2001. Google ScholarCross Ref
- A. Bagnall and J. Lines. An experimental evaluation of nearest neighbour time series classification. arXiv preprint arXiv:1406.4757, 2014.Google Scholar
- D. J. Berndt and J. Clifford. Using dynamic time warping to find patterns in time series. In Knowledge Discovery in Databases: Papers from the 1994 AAAI Workshop, pages 359--370, 1994.Google Scholar
- K. Buza, A. Nanopoulos, and L. Schmidt-Thieme. Insight: efficient and effective instance selection for time-series classification. In Advances in Knowledge Discovery and Data Mining, pages 149--160. Springer, 2011. Google ScholarDigital Library
- F. Itakura. Minimum prediction residual principle applied to speech recognition. Acoustics, Speech and Signal Processing, IEEE Transactions on, 23(1):67--72, 1975.Google Scholar
- E. Keogh and C. A. Ratanamahatana. Exact indexing of dynamic time warping. Knowledge and information systems, 7(3):358--386, 2005.Google Scholar
- E. Keogh, Q. Zhu, B. Hu, Y. Hao, X. Xi, L. Wei, and C. A. Ratanamahatana. The UCR time series classification/clustering homepage. http://www.cs.ucr.edu/~eamonn/time series data/.Google Scholar
- S.-W. Kim, S. Park, and W. W. Chu. Efficient processing of similarity search under time warping in sequence databases: an index-based approach. Information Systems, 29(5):405--420, 2004. Google ScholarDigital Library
- S. MacLean and G. Labahn. Elastic matching in linear time and constant space. In Proc., Ninth IAPR Int'l. Workshop on Document Analysis Systems, pages 551--554, 2010.Google Scholar
- M. Müller. Dynamic time warping. In Information Retrieval for Music and Motion, pages 69--84. Springer Berlin Heidelberg, 2007.Google ScholarCross Ref
- C. A. Ratanamahatana and E. Keogh. Everything you know about dynamic time warping is wrong. In Third Workshop on Mining Temporal and Sequential Data, pages 22--25, 2004.Google Scholar
- C. A. Ratanamahatana and E. Keogh. Making time-series classification more accurate using learned constraints. In In proc. of SDM Int'l Conf, pages 11--22, 2004.Google ScholarCross Ref
- A. Raza. RTW supplementary content page. https://sites.google.com/site/randomizedtw/, 2015.Google Scholar
- H. Sakoe and S. Chiba. Dynamic programming algorithm optimization for spoken word recognition. Acoustics, Speech and Signal Processing, IEEE Transactions on, 26(1):43--49, 1978.Google Scholar
- S. Salvador and P. Chan. Toward accurate dynamic time warping in linear time and space. Intelligent Data Analysis, 11(5):561--580, 2007. Google ScholarDigital Library
- S. Spiegel, B.-J. Jain, and S. Albayrak. Fast time series classification under lucky time warping distance. In Proceedings of the 29th Annual ACM Symposium on Applied Computing, pages 71--78. ACM, 2014. Google ScholarDigital Library
- X. Xi, E. Keogh, C. Shelton, L. Wei, and C. A. Ratanamahatana. Fast time series classification using numerosity reduction. In Proceedings of the 23rd international conference on Machine learning, pages 1033--1040. ACM, 2006. Google ScholarDigital Library
Index Terms
- Trading off accuracy for efficiency by randomized greedy warping
Recommendations
Fast time series classification under lucky time warping distance
SAC '14: Proceedings of the 29th Annual ACM Symposium on Applied ComputingIn time series mining, the Dynamic Time Warping (DTW) distance is a commonly and widely used similarity measure. Since the computational complexity of the DTW distance is quadratic, various kinds of warping constraints, lower bounds and abstractions ...
Dynamic Time Warping under limited warping path length
Dynamic Time Warping (DTW) is probably the most popular distance measure for time series data, because it captures flexible similarities under time distortions. However, DTW has long been suffering from the pathological alignment problem, and most ...
Randomized greedy algorithms for covering problems
GECCO '18: Proceedings of the Genetic and Evolutionary Computation ConferenceGreedy algorithms provide a fast and often also effective solution to many combinatorial optimization problems. However, it is well known that they sometimes lead to low quality solutions on certain instances. In this paper, we explore the use of ...
Comments