A Hybrid Multi-objective PSO-SA Algorithm for the Fuzzy Rule Based Classifier Design Problem with the Order Based Semantics of Linguistic Terms

Abstract

The studies in [26, 28, 33] has shown that the method of designing fuzzy rule based classification systems (FRBCSs) using multi-objective optimization evolutionary algorithms (MOEAs) clearly depends on evolutionary quality. For more specifically, the FRBCS design method using particle swarm optimization (PSO) is better than using genetic algorithm (GA). To prove that affirmation, this paper represents an application of a hybrid multi-objective particle swarm optimization algorithm with simulated annealing (MOPSO-SA) to optimize the semantic parameters of linguistic variables and fuzzy rule selection in designing FRBCSs based on hedge algebras proposed in [7] (using GSA-genetic simulated annealing algorithm). By simulation, MOPSO-SA is shown to be more efficient and produces better results than GSA-algorithm in [7] and the multi-objective PSO algorithm with fitness sharing (MO-PSO) in [33]. That is to show a method of the FRBCS design is better than another one using MOEA, the same MOEA must be used.