Optimization of fuzzy classifier parameters with a combination of gravitational search algorithm and shuffled frog leaping algorithm

In the present article, we analyse the effectiveness of combining two metaheuristic algorithms for tuning parameters of a fuzzy classifier. To work with imbalanced data, a fitness function is used based on a compromise between the overall accuracy and the geometric mean of accuracy of each class. The experiment was performed on data sets from the “Knowledge Extraction based on Evolutionary Learning” repository with different imbalance coefficients.