A New Algorithm for Online Management of Fuzzy Rules Base for Nonlinear Modeling
In this paper a new algorithm for online management of fuzzy rules base for nonlinear modeling is proposed. The online management problem is complex due to limitations of memory and time needed for calculations. The proposed algorithm allows an online creation and management of fuzzy rules base. It is distinguished, among the others, by mechanisms of: managing of number of fuzzy rules, managing of fuzzy rules weights and possibilities of background learning. The proposed algorithm was tested on typical nonlinear modeling problems.
KeywordsNeuro-fuzzy system Evolving fuzzy system Background learning
The project was financed by the National Science Centre (Poland) on the basis of the decision number DEC-2012/05/B/ST7/02138.
- 3.Kaisler, S., Armour, F., Espinosa, J.A., Money, W.: Big data: issues and challenges moving forward, system sciences (HICSS). In: Proceedings of the 46th Hawaii International Conference on, IEEE, pp. 995–1004 (2013)Google Scholar
- 4.Kibler, D., Aha, D.: Instance-based prediction of real-valued attributes. In Proceedings of the CSCSI (Canadian AI) Conference (1988)Google Scholar
- 8.Lughofer, E., Hüllermeier, E.: On-line redundancy elimination in evolving fuzzy regression models using a fuzzy inclusion measure. In: Proceedings of the EUSFLAT 2011 Conference, Elsevier, Aix-Les-Bains, France, pp. 380–387 (2011)Google Scholar
- 11.Rutkowski, L., Cpałka, K.: Flexible weighted neuro-fuzzy systems. In: Proceedings of the 9th International Conference on Neural Information Processing (ICONIP’02), Orchid Country Club, Singapore, November 18–22 (2002)Google Scholar
- 12.Rutkowski, L., Cpałka, K.: Neuro-fuzzy systems derived from quasi-triangular norms. In: IEEE International Conference on Fuzzy Systems, Budapest, July 26–29, vol. 2, pp. 1031–1036 (2004)Google Scholar
- 17.Waugh, S.: Extending and benchmarking Cascade-Correlation. PhD thesis, Computer Science Department, University of Tasmania (1995)Google Scholar