SIFT-SS: An Advanced Steady-State Multi-Objective Genetic Fuzzy System
Nowadays, automatic learning of fuzzy rule-based systems is being addressed as a multi-objective optimization problem. A new research area of multi-objective genetic fuzzy systems (MOGFS) has capture the attention of the fuzzy community. Despite the good results obtained, most of existent MOGFS are based on a gross usage of the classic multi-objective algorithms. This paper takes an existent MOGFS and improves its convergence by modifying the underlying genetic algorithm. The new algorithm is tested in a set of real-world regression problems with successful results.
Unable to display preview. Download preview PDF.
- 1.Ishibuchi, H.: Multiobjective genetic fuzzy systems: Review and future research directions. In: Proc. of 2007 IEEE International Conference on Fuzzy Systems, London, UK, July 23-26, pp. 913–918 (2007)Google Scholar
- 2.Ishibuchi, H.: Evolutionary multiobjective design of fuzzy rule-based systems. In: Proc. of 2007 IEEE Symposium on Foundation of Computational Intelligence, Honolulu, USA, April 1-5, pp. 9–16 (2007)Google Scholar
- 4.Ishibuchi, H., Tsukamoto, N., Nojima, Y.: Evolutionary Many-Objective Optimization. In: Proc. of 3rd International Workshop on Genetic and Evolving Fuzzy Systems, Witten-Bommerholz, Germany, pp. 47–52 (2008)Google Scholar
- 6.Casillas, J.: Efficient multi-objective genetic tuning of fuzzy models for large-scale regression problems. In: Proc. of 2009 IEEE International Conference on Fuzzy Systems, Jeju, Republic of Korea, pp. 1712–1717 (2009)Google Scholar
- 7.Van Veldhuizen, D.A.: Multiobjective Evolutionary Algorithms: Classifications, Analyses, and New Innovations. Ph.D. dissertation. Air Force Institute of Technology, Dayton (1999)Google Scholar
- 9.Demšar, J.: Statistical Comparisons of Classifiers over Multiple Data Sets. Journal of Machine Learning Research 7, 30 (2006)Google Scholar