Journal of Global Optimization
Introduction to Evolutionary Computing
Introduction to Evolutionary Computing
Design of Modern Heuristics: Principles and Application
Design of Modern Heuristics: Principles and Application
IEEE Transactions on Fuzzy Systems
Hi-index | 0.00 |
This paper suggests a genetic participatory learning algorithm and illustates its use in fuzzy systems modeling. The algorithm emerges from the concepts of participatory learning, selective transfer, and differential evolution. In genetic participatory learning the current population plays an important role in shaping evolution of the population individuals themselves. Selection uses compatibility between best and ramdonly chosen individuals. Exchange of information between individuals employes a recombination operator built from a selective transfer mechanism, whereas mutation proceeds analogously as in differential evolution. Recombination and mutation operations are affected by compatibility between individuals. An application example regarding fuzzy modeling of an electric maintenance problem using actual data serves to illustrate the effectveness of the algorithm, and to compare with alternative participatory and genetic fuzzy systems approaches. Computational results suggest that genetic participatory learning produces accurate and competitive models when compared with current state of the art approaches.