Comparative analysis of the similarity measures based on the moving approximation transformation in problems of time series classification
https://doi.org/10.15514/ISPRAS-2016-28(6)-15
Abstract
One of the major issues dealing with time-series classification problem is the choice of similarity measure. This article presents a comparative analysis of the similarity measure for time series based on moving approximations transform (MAP transforms) with other two most useful measures: Algorithm Dynamic Transformation and Euclidean distance for classification task. In addition, algorithm, that improves the precision of the measure for time series, that have similar values, but shifted relative to each other on the axis X, where coordinate on the X axis represents the time unit, is proposed.
About the Authors
I. S. Alimova
Kazan Federal University
Russian Federation
Russian Federation
V. D. Solovyev
Kazan Federal University
Russian Federation
Russian Federation
I. Z. Batyrshin
Instituto Politecnico Nacional
Russian Federation
Russian Federation
References
1. Weiss S. M. A novel approximation to dynamic time warping allows anytime clustering of massive time series datasets. Proceedings of the 2012 SIAM International Conference on Data Mining, 2012, pp. 999-1010. DOI: 10.1137/1.9781611972825.86.
2. Giusti R., Batista G. E. An empirical comparison of dissimilarity measures for time series classification. Intelligent Systems (BRACIS), 2013 Brazilian Conference on. – IEEE, 2013, pp. 82-88.
3. Ding H. et al. Querying and mining of time series data: experimental comparison of representations and distance measures. Proceedings of the VLDB Endowment, 2015, vol. 1, issue 2, pp. 1542-1552.
4. Batyrshin, I., Herrera-Avelar, R., Sheremetov, L., & Suarez, R. Moving approximations in time series data mining. Proc. Int. Conf. Fuzzy Sets and Soft Computing in Economics and Finance FSSCEF, 2004, pp. 62-72.
5. Almanza V., Batyrshin I. On trend association analysis of time series of atmospheric pollutants and meteorological variables in Mexico City Metropolitan Area. Mexican Conference on Pattern Recognition. Springer Berlin Heidelberg, 2011, pp. 95-102.
6. Batyrshin I.Z., Koshulski A.1, Sheremetov L.B.2, Klimova A.S.3, Panova A.M.4. Oil wells interaction analysis based on hybrid clustering of wells productivity time series. Nechetkie sistemy i mjagkie vychislenija [Fuzzy Systems and Soft Computations]. Tverskoj gosudarstvennyj universitet [Tver State University], 2007, vol. 2, issue 4, pp. 63-73 (in Russian).
7. E. Keogh, X. Xi, L. Wei, and C. A. Ratanamahatana. (2006) The UCR time series classification/clustering homepage. Available at http://www.cs.ucr.edu/~eamonn/time_series_data.
8. M. Muller. Dynamic time warping. Inf. Retr. Music Motion. Information retrieval for music and motion. Springer, Berlin, 2007, pp. 69–84.
9. Lu G. et al. A novel framework of change-point detection for machine monitoring. Mechanical Systems and Signal Processing, 2017, vol. 83, pp. 533-548.
10. Rath T. M., Manmatha R. Word image matching using dynamic time warping. Computer Vision and Pattern Recognition. Proceedings IEEE Computer Society Conference on, 2003, vol. 2, pp. 521-527.
11. Muda L., Begam M., Elamvazuthi I. Voice recognition algorithms using mel frequency cepstral coefficient (MFCC) and dynamic time warping (DTW) techniques. Journal of Computing, 2010, vol. 2, issue 3, pp. 138-143.
12. Vakanski A. et al. Trajectory learning for robot programming by demonstration using hidden Markov model and dynamic time warping. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2012, vol. 42, issue 4, pp. 1039-1052.
13. Keogh E. J., Pazzani M. J. Derivative Dynamic Time Warping .Sdm, 2001, vol. 1, pp. 5-7.
14. Faloutsos C., Ranganathan M., Manolopoulos Y. Fast subsequence matching in time-series databases. Proceedings of the 1994 ACM SIGMOD international Conference on Management of Data, 1994, vol. 23, issue 2, pp. 419-429.
Review
For citations:
Alimova I.S., Solovyev V.D., Batyrshin I.Z. Comparative analysis of the similarity measures based on the moving approximation transformation in problems of time series classification. Proceedings of the Institute for System Programming of the RAS (Proceedings of ISP RAS). 2016;28(6):207-222. (In Russ.) https://doi.org/10.15514/ISPRAS-2016-28(6)-15
Email this article 