Abstract
We suggest a simple yet effective and parameter-free feature construction process for time series classification. Our process is decomposed in three steps: (i) we transform original data into several simple representations; (ii) on each representation, we apply a coclustering method; (iii) we use coclustering results to build new features for time series. It results in a new transactional (i.e. object-attribute oriented) data set, made of time series identifiers described by features related to the various generated representations. We show that a Selective Naive Bayes classifier on this new data set is highly competitive when compared with state-of-the-art times series classification methods while highlighting interpretable and class relevant patterns.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
In contrast to lazy learning, eager learning has an explicit training phase and generally deploys faster.
- 2.
khc and snb are both available at http://www.khiops.com.
References
Bagnall, A., Davis, L.M., Hills, J., Lines, J.: Transformation based ensembles for time series classification. In: SDM’12, pp. 307–318 (2012)
Batal, I., Sacchi, L., Bellazzi, R., Hauskrecht, M.: Multivariate time series classification with temporal abstractions. In: FLAIRS’09 (2009)
Boullé, M.: A bayes optimal approach for partitioning the values of categorical attributes. J. Mach. Learn. Res. 6, 1431–1452 (2005)
Boullé, M.: MODL: a bayes optimal discretization method for continuous attributes. Mach. Learn. 65(1), 131–165 (2006)
Boullé, M.: Compression-based averaging of selective naive Bayes classifiers. J. Mach. Learn. Res. 8, 1659–1685 (2007)
Boullé, M.: Functional data clustering via piecewise constant nonparametric density estimation. Pattern Recogn. 45(12), 4389–4401 (2012)
Buza, K.A.: Fusion methods for time-series classification. Ph.D. thesis, University of Hildesheim (2011)
Demsar, J.: Statistical comparisons of classifiers over multiple data sets. J. Mach. Learn. Res. 7, 1–30 (2006)
Rodríguez, J.J., Alonso, C.J., Boström, H.: Learning first order logic time series classifiers: rules and boosting. In: Zighed, D.A., Komorowski, J., Żytkow, J.M. (eds.) PKDD 2000. LNCS (LNAI), vol. 1910, pp. 299–308. Springer, Heidelberg (2000)
Ding, H., Trajcevski, G., Scheuermann, P., Wang, X., Keogh, E.J.: Querying and mining of time series data: experimental comparison of representations and distance measures. PVLDB 1(2), 1542–1552 (2008)
Eruhimov, V., Martyanov, V., Tuv, E.: Constructing high dimensional feature space for time series classification. In: Kok, J.N., Koronacki, J., Lopez de Mantaras, R., Matwin, S., Mladenič, D., Skowron, A. (eds.) PKDD 2007. LNCS (LNAI), vol. 4702, pp. 414–421. Springer, Heidelberg (2007)
Geurts, P.: Pattern extraction for time series classification. In: Siebes, A., De Raedt, L. (eds.) PKDD 2001. LNCS (LNAI), vol. 2168, p. 115. Springer, Heidelberg (2001)
Grabocka, J., Nanopoulos, A., Schmidt-Thieme, L.: Invariant time-series classification. In: Flach, P.A., De Bie, T., Cristianini, N. (eds.) ECML PKDD 2012, Part II. LNCS, vol. 7524, pp. 725–740. Springer, Heidelberg (2012)
Grünwald, P.: The Minimum Description Length Principle. MIT Press, Cambridge (2007)
Hidasi, B., Gáspár-Papanek, C.: ShiftTree: an interpretable model-based approach for time series classification. In: Gunopulos, D., Hofmann, T., Malerba, D., Vazirgiannis, M. (eds.) ECML PKDD 2011, Part II. LNCS, vol. 6912, pp. 48–64. Springer, Heidelberg (2011)
Kadous, M.W., Sammut, C.: Classification of multivariate time series and structured data using constructive induction. Mach. Learn. 58(2–3), 179–216 (2005)
Keogh, E., Zhu, Q., Hu, B., Hao. Y., Xi, X., Wei, L., Ratanamahatana, C.A.: The UCR time series classification/clustering page (2011). http://www.cs.ucr.edu/~eamonn/time_series_data/
Liao, T.W.: Clustering of time series data - a survey. Pattern Recogn. 38(11), 1857–1874 (2005)
Lines, J., Davis, L.M., Hills, J., Bagnall, A.: A shapelet transform for time series classification. In: KDD’12, pp. 289–297 (2012)
Mörchen, F.: Time series feature extraction for data mining using DWT and DFT. Technical report, Philipps Univeristy Marburg (2003)
Nanopoulos, A., Alcock, R., Manolopoulos, Y.: Feature-based classification of time-series data. In: Mastorakis, N., Nikolopoulos, S.D. (eds.) Information Processing and Technology, pp. 49–61. Nova Science (2001)
Rakthanmanon, T., Keogh, E.: Fast shapelets: a scalable algorithm for discovering time series shapelets. In: SIAM DM’13 (2013)
Ramsay, J., Silverman, B.: Functional Data Analysis. Springer, New York (2005)
Shannon, C.E.: A mathematical theory of communication. Bell System Technical Journal (1948)
Stefan, A., Athitsos, V., Das, G.: The move-split-merge metric for time series. Trans. Knowl. Data Eng. 25, 1425–1438 (2013)
Vilalta, R., Rish, I.: A decomposition of classes via clustering to explain and improve naive bayes. In: Lavrač, N., Gamberger, D., Todorovski, L., Blockeel, H. (eds.) ECML 2003. LNCS (LNAI), vol. 2837, pp. 444–455. Springer, Heidelberg (2003)
Wang, X., Mueen, A., Ding, H., Trajcevski, G., Scheuermann, P., Keogh, E.: Experimental comparison of representation methods and distance measures for time series data. Data Min. Knowl. Disc. 26(2), 275–309 (2013)
Xi, X., Keogh, E.J., Shelton, C.R., Wei, L., Ratanamahatana, C.A.: Fast time series classification using numerosity reduction. In: ICML’06, pp. 1033–1040 (2006)
Xing, Z., Pei, J., Yu, P.S., Wang, K.: Extracting interpretable features for early classification on time series. In: SDM’11, pp. 247–258 (2011)
Yamada, Y., Suzuki, E., Yokoi, H., Takabayashi, K.: Decision-tree induction from time-series data based on a standard-example split test. In: ICML’03, pp. 840–847 (2003)
Ye, L., Keogh, E.J.: Time series shapelets: a novel technique that allows accurate, interpretable and fast classification. Data Min. Knowl. Disc. 22(1–2), 149–182 (2011)
Zhang, H., Ho, T.-B., Lin, M.-S.: A non-parametric wavelet feature extractor for time series classification. In: Dai, H., Srikant, R., Zhang, Ch. (eds.) PAKDD 2004. LNCS (LNAI), vol. 3056, pp. 595–603. Springer, Heidelberg (2004)
Acknowledgments
We wish to thank Anthony Bagnall and his team from University of East-Anglia for providing tsc-ensemble prototype, Eamonn Keogh and his team from University of California Riverside for providing prototypes of dtw-nn and fast-shapelets.
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Gay, D., Guigourès, R., Boullé, M., Clérot, F. (2014). Feature Extraction over Multiple Representations for Time Series Classification. In: Appice, A., Ceci, M., Loglisci, C., Manco, G., Masciari, E., Ras, Z. (eds) New Frontiers in Mining Complex Patterns. NFMCP 2013. Lecture Notes in Computer Science(), vol 8399. Springer, Cham. https://doi.org/10.1007/978-3-319-08407-7_2
Download citation
DOI: https://doi.org/10.1007/978-3-319-08407-7_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-08406-0
Online ISBN: 978-3-319-08407-7
eBook Packages: Computer ScienceComputer Science (R0)