Learning Fuzzy Rules Using Ant Colony Optimization Algorithms

Within the Linguistic Modeling field, one of the most important applications of Fuzzy Rule-Based Systems, the automatic learning from numerical data of the fuzzy linguistic rules composing these systems is an important task. In this paper we introduce a novel way of addressing the problem making use of Ant Colony Optimization (ACO) algorithms. To do so, the learning task will be formulated as an optimization problem and the features necessary for an ACO algorithm will be introduced. The behavior of the proposed learning method will be analyzed, compared with other ones, when solving of two applications with different characteristics: a three-dimensional function and a real-world electric engineering problem.