Abstract
This paper presents a new algorithm for nonlinear prediction based on independent component analysis mixture modelling (ICAMM). The data are considered from several mutually-exclusive classes which are generated by different ICA models. This strategy allows linear local projections that can be adapted to partial segments of a data set while maintaining generalization (capability for nonlinear modelling) given the mixture of several ICAs. The resulting algorithm is a general purpose technique that could be applied to time series prediction, to recover missing data in images, etc. The performance of the proposed method is demonstrated by simulations in comparison with several classical linear and nonlinear methods.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Hyvärinen, A., Karhunen, J., Oja, E.: Independent Component Analysis. John Wiley & Sons, New York (2001)
Lee, T.W., Lewicki, M.S., Sejnowski, T.J.: ICA mixture models for unsupervised classification of non-gaussian classes and automatic context switching in blind signal separation. IEEE Trans. on Patt. Analysis and Mach. Intellig. 22(10), 1078–1089 (2000)
Malaroiu, S., Kiviluoto, K., Oja, E.: ICA Preprocessing for Time Series Prediction. In: 2nd International Workshop on ICA and BSS (ICA 2000), pp. 453–457 (2000)
Pajunen, P.: Extensions of Linear Independent Component Analysis: Neural and Information-Theoretic Methods. Ph.D. Thesis, Helsinki University of Technology (1998)
Gorriz, J.M., Puntonet, C.G., Salmeron, G., Lang, E.W.: Time Series Prediction using ICA Algorithms. In: Proceedings of the 2nd IEEE International Workshop on Intelligent Data Acquisit. and Advanc. Comput. Systems: Technology and Applications, pp. 226–230 (2003)
Wang, C.Z., Tan, X.F., Chen, Y.W., Han, X.H., Ito, M., Nishikawa, I.: Independent component analysis-based prediction of O-Linked glycosylation sites in protein using multi-layered neural networks. In: IEEE 10th Internat. Conf. on Signal Processing, pp. 1–4 (2010)
Zhang, Y., Teng, Y., Zhang, Y.: Complex process quality prediction using modified kernel partial least squares. Chemical Engineering Science 65, 2153–2158 (2010)
Salazar, A., Vergara, L., Serrano, A., Igual, J.: A general procedure for learning mixtures of independent component analyzers. Pattern Recognition 43(1), 69–85 (2010)
Bersektas, D.: Nonlinear programming. Athena Scientific, Massachusetts (1999)
Cardoso, J.F., Souloumiac, A.: Blind beamforming for non gaussian signals. IEE Proceedings-F 140(6), 362–370 (1993)
Salazar, A., Vergara, L., Llinares, R.: Learning material defect patterns by separating mixtures of independent component analyzers from NDT sonic signals. Mechanical Systems and Signal processing 24(6), 1870–1886 (2010)
Salazar, A., Vergara, L.: ICA mixtures applied to ultrasonic nondestructive classification of archaeological ceramics. EURASIP Journal on Advances in Signal Processing, vol. 2010, p.11, Article ID 12520111 (2010), doi:10.1155/2010/125201
Salazar, A., Vergara, L., Miralles, R.: On including sequential dependence in ICA mixture models. Signal Processing 90(7), 2314–2318 (2010)
Raghavan, R.S.: A Model for Spatially Correlated Radar Clutter. IEEE Trans. on Aerospace and Electronic Systems 27, 268–275 (1991)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Safont, G., Salazar, A., Vergara, L. (2011). Nonlinear Prediction Based on Independent Component Analysis Mixture Modelling. In: Cabestany, J., Rojas, I., Joya, G. (eds) Advances in Computational Intelligence. IWANN 2011. Lecture Notes in Computer Science, vol 6692. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-21498-1_64
Download citation
DOI: https://doi.org/10.1007/978-3-642-21498-1_64
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-21497-4
Online ISBN: 978-3-642-21498-1
eBook Packages: Computer ScienceComputer Science (R0)