Improvement to the cooperative rules methodology by using the ant colony system algorithm.

Título inglés	Improvement to the cooperative rules methodology by using the ant colony system algorithm.
Título español	Mejoras a la metodología de reglas cooperativas utilizando el algoritmo del sistema de colonias de hormigas.
Autor/es	Alcalá, Rafael ; Casillas, Jorge ; Cordón, Oscar ; Herrera, Francisco
Organización	Dep. Comput. Sci. Univ. Jaén, Jaén, España;Dep. Comput. Sci. Artif. Intel. Univ. Granada, Granada, España
Revista	1134-5632
Publicación	2001, 8 (3): 321-335, 16 Ref.
Tipo de documento	articulo
Idioma	Inglés
Resumen inglés	The cooperative rules (COR) methodology [2] is based on a combinatorial search of cooperative rules performed over a set of previously generated candidate rule consequents. It obtains accurate models preserving the highest interpretability of the linguistic fuzzy rule-based systems. Once the good behavior of the COR methodology has been proven in previous works, this contribution focuses on developing the process with a novel kind of metaheuristic algorithm: the ant colony system one. Thanks to the capability of this algorithm to include heuristic information, the learning process is accelerated without model accuracy losses. Its behavior is successful compared with other processes based on genetic algorithms and sinlulated annealing when solving two modeling applications.
Clasificación UNESCO	120304 ; 110208
Palabras clave español	Lógica difusa ; Algoritmos de aprendizaje ; Heurística
Código Z-Math	Zbl 0996.68243
Acceso al artículo completo