Abstract
This paper suggests an evolving granular min-max regression algorithm for fuzzy rule-based system modeling. The algorithm starts with an empty rule base, but adds or modifies the rule base as stream data are input. Granulation of data is done by partitioning the input space using hyperboxes and associating to each hyperbox a fuzzy set and a fuzzy functional rule with affine consequent. The model output is produced combining the affine consequents weighted by the normalized membership degrees of the active fuzzy rules. The parameters of the consequents are adjusted using the recursive least squares with a forgetting factor. The algorithm has an incremental nature, and learns with one-pass processing of the data. The recursive form of the algorithm allows gradual model changes using simple maximum, minimum, and comparison operations, an appealing feature when handling high-dimensional data. Computational experiments concerning time series forecasting and system identification show that the evolving granular fuzzy min-max algorithm is fast, memory efficient, and competitive with current state of the art approaches.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Angelov, P., Filev, D.: An approach to on-line identification of evolving Takagi-Sugeno fuzzy models. IEEE Trans. Syst. Man Cybern. Part B 34(1), 484–498 (2004)
Angelov, P., Zhou, X.: Evolving fuzzy systems from data streams in real-time. In: International Symposium on Evolving Fuzzy Systems, Ambleside, Lake District, UK, pp. 29–35. IEEE Press (2006)
Aström, K., Wittenmark, B.: Computer-Controlled Systems, 3rd edn. Prentice-Hall Inc., Upper Saddle River (1997)
Falah, M., Lim, C.: Improving the fuzzy min-max neural network with a K-nearest hyperbox expansion rule for pattern classification. Appl. Soft Comput. 52, 135–145 (2017)
Filev, D., Georgieva, O.: An extended version of the Gustafson-Kessel algorithm for evolving data stream clustering. In: Evolving Intelligent Systems: Methodology and Applications, pp. 273–299. Wiley, Hoboken (2010)
Gabrys, B., Bargiela, A.: General fuzzy min–max neural network for clustering and classification. IEEE Trans. Neural Netw. Inst. Electr. Electr. Eng. (IEEE) 11(3), 769–783 (2000)
Box, G., Jenkins, G.: Time Series Analysis: Forecasting and Control, 2nd edn. Holden Day, San Francisco (1976)
Lima, E., Hell, M., Ballini, R., Gomide, F.: Evolving fuzzy modeling using participatory learning. In: Angelov, P., Filev, D., Kasabov, N. (eds.) Evolving Intelligent Systems: Methodology and Applications. Wiley, Hoboken (2010)
Lughofer, E.: Evolving Fuzzy Systems – Methodologies. Advanced Concepts and Applications. Springer, Heidelberg (2011)
Ma, D., Liu, J., Wang, Z.: The pattern classification based on fuzzy min-max neural network with new algorithm. In: Wang, J., Yen, G., Polycarpou, M. (eds.) Proceedings of 9th International Conference on Advances in Neural Networks - Part II (ISNN 2012), pp. 1–9. Springer, Heidelberg (2012)
Mascioli, F., Martinelli, G.: A constructive approach to neuro-fuzzy networks. Sig. Process. 64(3), 347–358 (1998). Elsevier
Seera, M., Lim, C., Loo, C., Singh, H.: A modified fuzzy min–max neural network for data clustering and its application to power quality monitoring. Appl. Soft Comput. 28, 19–29 (2015). Elsevier
Simpson, P.: Fuzzy min–max neural networks part 1: classification. IEEE Trans. Neural Netw. 3(5), 776–786 (1992)
Simpson, P.: Fuzzy min–max neural networks part 2: clustering. IEEE Trans. Fuzzy Syst. 1(1), 32 (1993)
Tagliaferri, R., Eleuteri, A., Meneganti, M., Barone, F.: Fuzzy min–max neural networks: from classification to regression. Soft Comput. 5(1), 69–76 (2001). Springer
Zhang, H., Liu, J., Ma, D., Wang, Z.: Data-core-based fuzzy min–max neural network for pattern classification. IEEE Trans. Neural Netw. 22(12), 2339–2352 (2011)
Acknowledgement
The authors thank the Brazilian National Council for Scientific and Technological Development (CNPq) for a fellowship, and grant 305906/2014-3, respectively.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG
About this paper
Cite this paper
Porto, A., Gomide, F. (2018). Evolving Granular Fuzzy Min-Max Regression. In: Melin, P., Castillo, O., Kacprzyk, J., Reformat, M., Melek, W. (eds) Fuzzy Logic in Intelligent System Design. NAFIPS 2017. Advances in Intelligent Systems and Computing, vol 648. Springer, Cham. https://doi.org/10.1007/978-3-319-67137-6_18
Download citation
DOI: https://doi.org/10.1007/978-3-319-67137-6_18
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-67136-9
Online ISBN: 978-3-319-67137-6
eBook Packages: EngineeringEngineering (R0)