A Comparison Study of Fuzzy Transform based Approximation Methods
Hee-Jun Min, Hye-Young Jung · 2022 Joint 12th International Conference on Soft Computing and Intelligent Systems and 23rd International Symposium on Advanced Intelligent Systems (SCIS&ISIS) · 2022
Fuzzy Transform based on fuzzy set is a method for data compression and function approximation. Least Sqaures Fuzzy Transform that applied L2-norm to fuzzy transform has been proposed. Least Squares Fuzzy Transform shows good performance on function approximation, but is sensitive to outliers. Recently, Least Absolute Deviation Fuzzy Transform that applied L1-norm to fuzzy transform which is robust to outliers and function approximation has been proposed. In this paper, we compare the performance of three methods in terms of function approximation and outlier robustness problems. Experiments show that the Least Absolute Deviation Fuzzy Transform outperforms the others in function approximation and outlier robustness problems.