Publishing house Radiotekhnika

"Publishing house Radiotekhnika":
scientific and technical literature.
Books and journals of publishing houses: IPRZHR, RS-PRESS, SCIENCE-PRESS

Тел.: +7 (495) 625-9241


Identification of periodic patterns in time series


T.V. Afanasjeva – Dr. Sc. (Eng.), Professor, Associate Professor, Department «Information Systems»,
Ulyanovsk State Technical University
A.A. Sapunkov – Post-graduate Student, Ulyanovsk State Technical University
V.M. Stuchebnikov – Dr. Sc. (Eng.), Professor, General Director, JSC MIDAUS (Ulyanovsk)

Preprocessing and time series analysis is an important task in the field of process analysis. One of the tasks of the analysis is the task of detecting of events and their periods to simulate periodic and seasonal changes in time series. Known solutions rely on algorithms that are partly based on knowledge of the possible behavior of a time series or its model, which requires the involvement of an expert. However, when working with large volumes of data, the work of experts becomes less effective.
The article proposes the solution of the problem of identification of periodic patterns and their characteristics in linguistic time series. It is assumed that the numerical time series has been previously converted into linguistic one. Symbolic and segmental periodicity, represented in the form of linguistic patterns, repeated at regular intervals of time, are considered. The article introduces the definitions and statements that differentiate the notion of periodicity into symbolic and segmental. An algorithm for detecting periodicity in time series is proposed, which makes it possible to identify the parameters of repeating patterns with a constant period.
The proposed algorithm consists of two stages: the search for all repeating patterns in the linguistic time series and testing the pa-rameters of the generated patterns for the periodicity. The use of linguistic time series allows you to use a modified suffix tree (a pattern tree) to search for repeated patterns. Modifications of the algorithm for constructing this tree are aimed at obtaining the parameters of the patterns. At the second stage of the algorithm, each pattern is tested for repeatability with a constant period. The result of the proposed algorithm is the conclusion about the presence or absence of periodicity in the time series, as well as the parameters of the periodic patterns, which makes it possible to apply this information to their prediction.

  1. Malode Y.B., Khadse D.B., Jamthe D.V. Efficient Periodicity Mining using Circular Autocorrelation in TimeSeries Data // International Research Journal of Engineering and Technology (IRJET). 2015. P. 430−436.
  2. Rasheed F, Alhajj R. STNR: A suffix tree based noise resilient algorithm for periodicity detection in time series databases // Applied Intelligence. 2010. P. 267−278.
  3. Mala D, Mahanta A. Detection of calendar based periodicities of interval-based temporal patterns // International Journal of Data Mining & Knowledge Management Process (IJDKP). 2012. P. 17−31.
  4. Elfeky M.G., Aref W.G., Elmagarmid A.K. Periodicity Detection in Time Series Databases // IEEE Trans. Knowledge and Data Eng. 2005. P. 875−887.
  5. Elfeky M.G., Aref W.G., Elmagarmid A.K. WARP: Time Warping for Periodicity Detection // Proceedings of the Fifth IEEE International Conference on Data Mining (ICDM'05). 2005. P. 138−145.
  6. Novak V. Linguistic characterization of time series. // Fuzzy Sets and Systems. 2016. P. 52−72.
  7. Knut D. Algoritm Knuta–Morrisa–Pratta // Iskusstvo programmirovaniya na E'VM. M.: Mir. 1978. T. 3. S. 356.
  8. Afanas'eva T.V., Toneryan M.S. Primenenie algoritma Knuta–Morrisa–Pratta dlya resheniya problemy' identifikaczii periodicheskix lokal'ny'x tendencziй v nechetkix vremenny'x ryadax // VI ya Vserossiйskaya nauchno-prakticheskaya konf. Nechetkie sistemy' i myagkie vy'chisleniya–2014 (NSMV–2014). T. 1. SPb: Izd-vo LITMO. 2014. S. 12−18.
  9. Lin J., Keogh E., Lonardi S., Chiu B.Y. A symbolic representation of time series, with implications for streaming algorithms // DMKD. 2003. P. 2−11.
  10. Afanasieva T., Yarushkina N., Gyskov G. ACL-Scale as a Tool for Preprocessing of Many-Valued Contexts // Proc. of the Second International Workshop on Soft Computing Applications and Knowledge Discovery (SCAD 2016). 2016. P. 2−11.
  11. Afanas'eva T.V. Model' ACL-shkaly' dlya generaczii lingvisticheskix oczenok v prinyatii reshenij // Voprosy' sovremennoj nauki i praktiki. Universitet im. V.I. Vernadskogo. T. 2. Seriya «Texnicheskie nauki». Tambov. TGTU. 2008. № 4(14). S. 91−97.
  12. McCreight E.M.. A space-economical suffix tree construction algorithm // Journal of the ACM. 1976. P. 262−272.
June 24, 2020
May 29, 2020

© Издательство «РАДИОТЕХНИКА», 2004-2017            Тел.: (495) 625-9241                   Designed by [SWAP]Studio