Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Research › peer-review
Development and Research of the Time Series Prediction Method Based on Finite State Automaton. / Pavlova, Ulyana; Rakitskiy, Anton.
Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021. Institute of Electrical and Electronics Engineers Inc., 2021. p. 305-307 9455056 (Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Research › peer-review
}
TY - GEN
T1 - Development and Research of the Time Series Prediction Method Based on Finite State Automaton
AU - Pavlova, Ulyana
AU - Rakitskiy, Anton
N1 - Publisher Copyright: © 2021 IEEE.
PY - 2021/5/13
Y1 - 2021/5/13
N2 - In this article, we explore the possibility of using deterministic finite state machines to forecast real time series. The proposed method is based on a ten-headed automaton that allows recognizing multilinear sequences. This machine has been redesigned, modified, and implemented taking into account the time series. Theoretically, the resulting automaton takes into account the appearance of deviations in the data and assumes adaptation when the pattern changes. In addition to modification, the authors considered the possibility of machine failure by other forecasting methods (for example, forecasting methods based on archivers). The article briefly describes the basic version of the machine, as well as all applied modifications with the justification for their use. Each modification of the automaton is examined both on sequences similar to a multilinear pattern and on data of stochastic origin. In addition, the article presents the results of applying the proposed implementations for forecasting dollar exchange rates.
AB - In this article, we explore the possibility of using deterministic finite state machines to forecast real time series. The proposed method is based on a ten-headed automaton that allows recognizing multilinear sequences. This machine has been redesigned, modified, and implemented taking into account the time series. Theoretically, the resulting automaton takes into account the appearance of deviations in the data and assumes adaptation when the pattern changes. In addition to modification, the authors considered the possibility of machine failure by other forecasting methods (for example, forecasting methods based on archivers). The article briefly describes the basic version of the machine, as well as all applied modifications with the justification for their use. Each modification of the automaton is examined both on sequences similar to a multilinear pattern and on data of stochastic origin. In addition, the article presents the results of applying the proposed implementations for forecasting dollar exchange rates.
KW - data compression
KW - finite state automaton
KW - information theory
KW - machine learning
KW - prediction
KW - time series analysis
UR - http://www.scopus.com/inward/record.url?scp=85113786916&partnerID=8YFLogxK
U2 - 10.1109/USBEREIT51232.2021.9455056
DO - 10.1109/USBEREIT51232.2021.9455056
M3 - Conference contribution
AN - SCOPUS:85113786916
T3 - Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021
SP - 305
EP - 307
BT - Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021
Y2 - 13 May 2021 through 14 May 2021
ER -
ID: 34143420