Rule selection by Guided Elitism genetic algorithm in Fuzzy Min-Max classifier
Hadis Jalesiyan, Mahdi Yaghubi, T. Akbarzadeh · 2014
Rule-based classification with Neural Networks has high acceptance ability for noisy data, high accuracy and is preferable in data mining. In this paper, we use Fuzzy Min-Max (FMM) Neural Network. Nevertheless the — Curse of Dimensionality — problem also exists in this classifier. As a possible solution, in this paper the modified GA is adopted to minimize the number of features in the extracted rules. “Guided Elitism” strategy is used to create elitism in the population, based on information extracted from good individuals of previous generations. The main advantage of this data structure is that it maintains partial information of good solutions, which may otherwise be lost in the selection process. Five well-known benchmark problems are used to evaluate the performance of the proposed GEGA system; Results shows comparatively high accuracy and generally lower computational time.