IMPLEMENTATION OF FUZZY TIME SERIES IN FORECASTING ONLINE STUDENT LEARNING DEVELOPMENT

Abstract

Education has an important role in improving the quality of personality in intellectual and moral aspects. The COVID-19 pandemic has affected all human activities, one of which is in the field of education. With the stigma circulating, the learning system that was originally carried out face-to-face has changed to learning from home. Problems that arise in online learning are many changes that occur significantly, especially obstacles to students' understanding of the material and also students must master all subjects in school in accordance with the majors. Purpose: forecasting can be used as an effective tool in helping predict the development of online student learning in the coming month or period. Methods: In this study, the forecasting system that will be designed has several factors that affect the results. The Fuzzy Time Series method is the right method, because it does not require assumptions as forecasting using classical forecasting methods. In this research, forecasting using data from students of SMKN 3 Boyolangu Tulungagung class X majoring in TPM with a sample of 55 students. Results: the forecasting system for student learning development using the Fuzzy Time Series method can work well, namely by displaying the actual value and the results of the Fuzzy Time Series method (determining intervals, fuzzification, FLR, FLRG and defuzzification). The MAPE level is 0.404%. Conclusion: the system is declared accurate in forecasting because it has an error rate of less than 1%.