Abstract
A novel sampling optimization scheme is proposed for the online sequential extreme learning machine (OS-ELM) based on improved Gath-Geva (IGG) fuzzy segmentation. As is known that most of the complex systems are time-varying in nature whose dynamics varies with changes of internal and environmental factors. When OS-ELM is implemented to identify the time-varying dynamics, it select samples in the sampling pool for a certain period of time. Under such circumstance, samples representing system dynamics of different time span are mixed together thus the online representing ability of OS-ELM for current dynamics is deteriorated. To construct optimal sample pool for OS-ELM thus improve its representing ability for current system dynamics, in this study, time series of system output and relevant factors are online segmented by an IGG fuzzy segmentation approach. Time series are segmented as per dynamics characteristics such as mean value and variance. The changing points split up the time series into several segments and the changing points themselves represent changes in system dynamics. Samples within the same segment are considered as possessing similar characteristics. The OS-ELM selects samples from sampling pool which consists of samples with better representing ability for current dynamics. Furthermore, the membership degree and novelty degree are both employed to decide the importance of samples thus improve the generalization ability of OS-ELM . To achieve accurate tidal prediction, the IGG-based sampling scheme is applied in online tidal prediction for representing the influences of environmental factors. In the meantime, conventional harmonic analysis is applied to represent the influences of celestial bodies and coastal topology. The harmonic method and improved OS-ELM are combined together and the resulted modular prediction scheme is applied for online tidal level prediction of port of King Point. Simulation results demonstrates the feasibility and effectiveness of the proposed sampling scheme and the modular tidal prediction approach.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Fang, G.H., Zheng, W.Z., Chen, Z.Y.: Analysis and Prediction of Tide and Tidal Current. Ocean Press, Beijing (1986)
Lee, T.: Back-propagation neural network for long-term tidal predictions. Ocean Eng. 31, 225–238 (2004)
Liang, S., Li, M., Sun, Z.: Prediction models for tidal level including strong meteorologic effects using a neural network. Ocean Eng. 35, 666–675 (2008)
Yin, J.C., Zou, Z.J., Xu, F.: Sequential learning radial basis function network for real-time tidal level predictions. Ocean Eng. 57, 49–55 (2013)
Liang, N.Y., Huang, G.B., Saratchandran, P., Sundararajan, N.: A fast and accurate online sequential learning algorithm for feedforward networks. IEEE Trans. Neur. Netw. 17(6), 1411–1423 (2006)
Huang, G.B., Zhu, Q.Y., Siew, C.K.: Extreme learning machine: theory and applications. Neurocomputing 70, 489–501 (2006)
Huang, G.B., Wang, D.H., Lan, Y.: Extreme learning machines: a survey. Int. J. Mach. Learn. Cyb. 2(2), 107–122 (2011)
Wang, N.N., Liu, X.D., Yin, J.C.: Improved Gath-Geva clustering for fuzzy segmentation of hydrometeorological time series. Stoch. Env. Res. Risk A. 26(1), 139–155 (2012)
Jiang, N., Zhao, Z.Y., Ren, L.Q.: Design of structural modular neural networks with genetic algorithm. Adv. Eng. Softw. 34, 17–24 (2003)
Huang, G.B., Ding, X.J., Zhou, H.M.: Optimization method based extreme learning machine for classification. Neurocomputing 74, 155–163 (2010)
Huang, G.B., Zhou, H.M., Ding, X.J., Zhang, R.: Extreme learning machine for regression and multiclass classification. IEEE Trans. Syst. Man Cy. B 42(2), 513–529 (2012)
Abonyi, J., Feil, B., Nemeth, S., Arva, P.: Fuzzy clustering based segmentation of time-series. In: Berthold, M., Lenz, H.-J., Bradley, E., Kruse, R., Borgelt, C. (eds.) IDA 2003. LNCS, vol. 2810, pp. 275–285. Springer, Heidelberg (2003)
Povinelli, R., Johnson, M., Lindgren, A., Ye, J.: Time series classification using Gaussian mixture models of reconstructed phase spaces. IEEE Trans. Knowl. Data Eng. 16(6), 779–783 (2004)
Figueiredo, M., Jain, A.K.: Unsupervised learning of finite mixture models. IEEE Trans. Pattern Anal. Mach. Intell. 24(3), 381–396 (2002)
Fu, Z., Robles-Kelly, A., Zhou, J.: Mixing linear SVMs for nonlinear classification. IEEE Trans. Neural Netw. 21, 1963–1975 (2010)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Yin, J., Wang, N. (2015). An Online Sequential Extreme Learning Machine for Tidal Prediction Based on Improved Gath-Geva Fuzzy Segmentation. In: Cao, J., Mao, K., Cambria, E., Man, Z., Toh, KA. (eds) Proceedings of ELM-2014 Volume 2. Proceedings in Adaptation, Learning and Optimization, vol 4. Springer, Cham. https://doi.org/10.1007/978-3-319-14066-7_24
Download citation
DOI: https://doi.org/10.1007/978-3-319-14066-7_24
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-14065-0
Online ISBN: 978-3-319-14066-7
eBook Packages: EngineeringEngineering (R0)