A Modified Pittsburg Approach to Design a Genetic Fuzzy Rule-Based Classifier from Data

摘要

The paper presents a modification of the Pittsburg approach to design a fuzzy classifier from data. Original, non-binary crossover and mutation operators are introduced. No special coding of fuzzy rules and their parameters is required. The application of the proposed technique to design the fuzzy classifier for the well known benchmark data set ( Wisconsin Breast Cancer) available from the http://archive.ics.uci.edu/ml is presented. A comparative analysis with several alternative (fuzzy) rule-based classification techniques has also been carried out.