A QUADRATIC MARGIN-BASED MODEL FOR WEIGHTING FUZZY CLASSIFICATION RULES INSPIRED BY SUPPORT VECTOR MACHINES
Mohammad Taheri, Hamid Azad, Koorush Ziarati, Reza Sanaye · 2013
Recently, tuning the weights of the rules in Fuzzy Rule-Base Clas- sification Systems is researched in order to improve the accuracy of classifica- tion. In this paper, a margin-based optimization model, inspired by Support Vector Machine classifiers, is proposed to compute these fuzzy rule weights. This approach not only considers both accuracy and generalization criteria in a single objective function, but also is independent of any order in presenting data patterns or fuzzy rules. It has a global optimum solution and needs only one regularization parameter C to be adjusted. In addition, a rule reduction method is proposed to eliminating low weighted rules and having a compact rule-base. This method is compared with some greedy, reinforcement and lo- cal search rule weighting methods on 13 standard datasets. The experimental results show that, the proposed method significantly outperforms the other ones especially from the viewpoint of generalization.