Generating a Fuzzy Rule Based Classification System by genetic learning of granularity level using TOPSIS
Antonio Eloy de Oliveira Araujo, Renato Antonio Krohling · 2019
Fuzzy Rule Based Classification Systems (FRBCSs) are widely used tools in classification problems.An important aspect in the design of a FRBCS is the number of fuzzy labels per variable (granularity level), which significantly influences the performance of the fuzzy system.Another relevant issue to be considered when generating a FRBCS is the accuracy-interpretability tradeoff, which can be addressed in the context of multiobjective optimization.Thus, in this work, we propose a new approach to design a FR-BCS in which the accuracy and the interpretability (number of rules) of the FRBCS are considered objectives to be treated with the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS).We applied our method to several well-known standard classification datasets and the results show the feasibility of the proposed approach.