An alternative fuzzy time series forecast model combined generalized fuzzy logical relationship, natural partitioning, and adaptive defuzzification

This paper, we propose generalized fuzzy logical relationships based on natural partitioning and adaptive defuzzification. The proposed method provides a better approach to improve performance by producing a good evaluation of the forecasted value. This study aims to minimize forecasting errors for each data series. The general suitability of the proposed model was tested by implementing it in the fore- casting of student enrollments at the University of Alabama. In order to show the superiority of the proposed model over existing methods, the results obtained have been compared with evaluations such as MSE, RMSE, MAE, MAPE, and forecasting error errors for each time series data. Comparative studies show that the proposed method is superior to existing methods for all evaluations provided.