Exponential Membership Function Evaluation based on Frequency

The unknown parameters of membership function may be a problem in fuzzy works. In that case, frequency table of data set may be a key in the determination of membership function. Considering this, in the current study, the parameter formulas of exponential membership function are generated via a minimization problem in where the frequency table structures are used. The offered exponential membership function is performed on a classification problem of bispectral index data sets in order to observe whether they are more effective than previous one used in the literature. At the end of the analysis, it is concluded that, the mean of classification accuracies based on offered exponential membership functions is statistically greater than other one.