BUILDING OF COMPACT AND ACCURACY FUZZY MODELS BASED ON STATISTICAL INFORMATION CRITERIA
This paper focuses on fuzzy models design using the statistical information criteria: 1) the Akaike information criterion; 2) the Bayes information criterion; and 3) the Hannan–Quinn information criterion. The training procedure is based on optimization and differential evolution. The performed analysis is based on ideal and real dataset showing the effectiveness of the proposed method.
Keywords: fuzzy rule-based model, ant colony optimization, differential evolution algorithm, Akaike information criterion, Bayes information criterion, Hannan–Quinn information criterion
