Accurate Learning Approach to Linguistic Model
Xin Jiang · Jisuanji fangzhen · 2006
Linguistic model is highly descriptive,but it suffers from inaccuracy in some complex problems.The respective characteristics of genetic algorithm and particle swarm algorithm are used to build a two-stage evolvement strategy for learning linguistic model meting accurate requirement.Particle swarm algorithm is used to optimize each membership function of linguistic terms form each variable.Candidate rule bases is incorporated and formed with linguistic terms.Genetic algorithm is utilized to select rules form the bases.The method requests scarcely any previous information and obtains fuzzy model from samples.The validity of the method has been demonstrated by examples of function approximation problem.