Informatics and Applications
2019, Volume 13, Issue 1, pp 40-48
LOCAL APPROXIMATION MODELS FOR HUMAN PHYSICAL ACTIVITY CLASSIFICATION
- D. A. Anikeyev
- G. O. Penkin
- V. V. Strijov
Abstract
The research is devoted to the time series classification. The time series is measured by an accelerometer of a wearable device. A class of physical activity is defined by its feature description of a time segment. To construct this description, the authors propose to use parameters of various approximation splines (algebraic, smoothing, adaptive regression, or spline with dynamic nodes). The logistic regression is used as a classifier. It delivers desired quality of the activity recognition. The authors analyze the space of the local approximation parameters. Classification accuracy depends on the method of this space construction. The computational experiment finds the optimal approximation parameters and parameters of the classifier.
[+] References (16)
- Farmer, J. D., and J. J. Sidorowich. 1987. Predicting chaotic time series. Phys. Rev. Lett. 59(8):845-848.
- Karasikov, M. E., and V.V. Strijov. 2008. Klassifikatsiya vremennykh ryadov v prostranstve parametrov porozh- dayushchikh modeley [Feature-based time-series classi-fication]. Informatika i ee Primeneniya - Inform. Appl. 10(4):128-138.
- Myasnikov, L. L., and E. N. Myasnikova. 1970. Av tomaticheskoe raspoznavanie zvukovykh obrazov [Automatic speech recognition]. Moscow: Energiya. 183 p.
- Kwapisz, J. R., G. M. Weiss, and S. A. Moore. 2011. Activity recognition using cell phone accelerometers. ACM SigKDD Explorations Newsletter 12(2):72-82.
- Kuznetsov, M.P, and N. P Ivkin. 2015. Algoritm klassifikatsii vremennykh ryadov akselerometra po kombinirovannomu priznakovomu opisaniyu [Time series classification algorithm using combined feature description].
Mashinnoe obuchenie ianalizdannykh [Machine Learning Data Anal.] 1(11):1471-1483.
- Tsay, R. S. 2010. Analysis of financial time series. Hoboken, NJ: Wiley. 712 p.
- Fu, C. 2011. A review on time series data mining. Eng. Appl.Artif. Intel. 24:164-181.
- Esling, P., and C. Agon. 2012. Time series data mining. ACM Comput. Surv. 45(1):1-34.
- Coull, B.A., and J. Staundenmayer. 2004. Self modeling regression for multivariate curve data. Stat. Sinica 14:695-711.
- Weber, G.-W., I. Batmaz, G. Koksal, etal. 2012. CMARS: A new contribution to nonparametric regression with mul- tivariative adaptive regression splines supported by contin-uous optimisation. Inverse Probl. Sci. En. 20(3):371-400. doi: 10.1080/17415977.2011.624770.
- Istomin, I. A., O. L. Kotlyarov, and A. Yu. Loskutov 2005. The problem of processing time series: Extending possi-bilities of the local approximation method using singular spectrum analysis. Theor. Math. Phys. 142(1):128-137.
- Dette, H., V. B. Melas, and A. Pepelyshev. 2011. Optimal design for smoothing splines Ann. I. Stat. Math. 63(5):981-1003.
- Tselykh, V. R. 2012. Mnogomernye adaptivnye regres- sionnye splayny [Multivariate adaptive regression splines]. Mashinnoeobuchenie ianalizdannykh [Machine Learning Data Anal.] 1(3):272-278.
- Gholizadel, S., and E. Salajegheh. 2009. Optimal design of structures subjected to time history loading by swarm in
telligence and an advanced metamodels. Comput. Method. Appl. M. 198:2936-2949.
- Vasko, K., and H. Toivonen. 2002. Estimating the number of segments in time series data using permutation tests. Conference (International) on Data Mining Proceedings 466-473.
- Motrenko, A. P., and V. V. Strijov. 2016. Extracting fun-damental periods to segment biomedical signals. IEEE J. Biomed. Health 20(6):1466-1476.
[+] About this article
Title
LOCAL APPROXIMATION MODELS FOR HUMAN PHYSICAL ACTIVITY CLASSIFICATION
Journal
Informatics and Applications
2019, Volume 13, Issue 1, pp 40-48
Cover Date
2019-04-30
DOI
10.14357/19922264190106
Print ISSN
1992-2264
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
local approximation model; time series; classification; splines; feature space
Authors
D. A. Anikeyev  , G. O. Penkin ,
and V. V. Strijov , 
Author Affiliations
Moscow Institute of Physics and Technology, 9 Institutskiy Per., Dolgoprudny, Moscow Region 141700, Russian Federation
A. A. Dorodnicyn Computing Center, Federal Research Center "Computer Science and Control" of Russian Academy of Sciences, 40 Vavilov Str., Moscow 119333, Russian Federation
|