Abstract
Time series discord has proved to be a useful concept for time series anomaly detection. To search for discords, various algorithms have been developed. HOT SAX has been considered as a well-known and effective algorithm in time series discord discovery. However this algorithm still has some weaknesses. First, users of HOT SAX are required to choose suitable values for the discord length, word-length and/or alphabet-size, which are unknown. Second, HOT SAX still suffers from high computation cost. In this paper, we propose some novel techniques to improve HOT SAX algorithm. These techniques consist of (i) using some time series segmentation methods to estimate the two important parameters: discord length and word length and (ii) speeding up the discord discovery process by a new way of shifting the sliding window. Extensive experiments have demonstrated that the proposed approach can not only facilitate users in setting the parameters, but also improve the discord discovery in terms of accuracy and computational efficiency.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bu, Y., Leung, T.W., Fu, A., Keogh, E., Pei, J., Meshkin, S.: WAT: Finding top-K discords in time series database. In: Proceedings of the 2007 SIAM International Conference on Data Mining (SDM 2007), Minneapolis, MN, USA, 26–28 April 2007
Dani, M.C., Jollois, F.X., Nadif, M., Freixo, C.: Adaptive threshold for anomaly detection using time series segmentation. In: Arik, S., Huang, T., Lai, W.K., Liu, Q. (eds.) ICONIP 2015, Part III. LNCS, vol. 9491, pp. 82–89. Springer, Switzerland (2015)
Fu, T.C., Chung, F.L., Ng, C.M.: Financial time series segmentation based on specialized binary tree representation. In: Proceedings of 2006 International Conference on Data Mining, pp. 3–9 (2006)
Gruber, C., Coduro, M., Sick, B.: Signature verification with dynamic RBF network and time series motifs. In: Proceedings of 10th International Workshop on Frontiers in Hand Writing Recognition (2006)
Keogh, E., Selina, C., David, H., Michel, P.: An online algorithm for segmenting time series. In: Proceedings of the IEEE International Conference on Data Mining, pp. 289–296 (2001)
Keogh, E., Lin, J., Fu, A.: HOT SAX: efficiently finding the most unusual time series subsequence. In: Proceedings of 5th ICDM, Houston, Texas, pp. 226–233 (2005)
Keogh, E.: www.cs.ucr.edu/~eamonn/discords/. (Accessed on 24 Jan 2015)
Kha, N.H., Anh, D.T.: From cluster-based outlier detection to time series discord discovery. In: Li, X.L., Cao, T., Lim, E.-P., Zhou, Z.-H., Ho, T.-B., Cheung, D. (eds.) PAKDD 2015. LNCS (LNAI), vol. 9441, pp. 16–28. Springer, Switzerland (2015)
Leng, M., Chen, X., Li, L.: Variable length methods for detecting anomaly patterns in time series. In: International Symposium on. Computational Intelligence and Design (ISCID 2008), vol. 2 (2008)
Li, G., Braysy, O., Jiang, L., Wu, Z., Wang, Y.: Finding time series discord based on bit representation clustering. Knowl.-Based Syst. 52, 243–254 (2013)
Lin, J., Keogh, E., Lonardi, S., Chiu, B.: Symbolic representation of time series, with implications for streaming algorithms. In: Proceedings of the 8th ACM SIGMOD Workshop on Research Issues in Data Mining and Knowledge Discovery, San Diego, CA, 13 June 2003
Lovric, M., Milanovic, M., Stamenkovic, M.: Algorithmic methods for segmentation of time series: an overview. JCEBI 1(1), 31–53 (2014)
Oliveira, A.L.I., Neto, F.B.L., Meira, S.R.L.: A method based on RBF-DAA neural network for improving Novelty detection in time series. In: Proceedings of 17th International FLAIRS Conference. AAAI Press, Miami Beach (2004)
Pratt, K.B., Fink, E.: Search for patterns in compressed time series. Int. J. Image Graph. 2(1), 89–106 (2002)
Salvador, S., Chan, P.: Learning states and rules for time series anomaly detection. Appl. Intell. 23(3), 241–255 (2005)
Tanaka, Y., Iwamoto, K., Uehara, K.: Discovery of time series motif from multi-dimensional data based on MDL principle. Mach. Learn. 58, 269–300 (2005)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering
About this paper
Cite this paper
Thuy, H.T.T., Anh, D.T., Chau, V.T.N. (2016). Some Efficient Segmentation-Based Techniques to Improve Time Series Discord Discovery. In: Vinh, P., Barolli, L. (eds) Nature of Computation and Communication. ICTCC 2016. Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, vol 168. Springer, Cham. https://doi.org/10.1007/978-3-319-46909-6_17
Download citation
DOI: https://doi.org/10.1007/978-3-319-46909-6_17
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-46908-9
Online ISBN: 978-3-319-46909-6
eBook Packages: Computer ScienceComputer Science (R0)