THE NAÏVE BAYES METHOD AS A MEASUREMENT MODEL EFFECTIVENESS OF ONLINE LEARNING

Abstract

The rapid development of technology requires the world of education to be able to take advantage of its positive impact, making various new innovations by utilizing technology to support education such as online learning in the learning process amid the Covid-19 pandemic. Changes in learning methods which occur suddenly from conventional learning methods or directly face-to-face switching to distance learning methods or using online learning media greatly impact and influence students who come from underprivileged families and students who are in remote areas where internet access and inadequate infrastructure. This study aims to create a classification model for measuring the effectiveness of online learning in Pringsewu using the classification method. The classification method is used to classify data based on the nature of the data which each class already recognizes. There are various methods which can be used to classify data using the Naïve Bayes method. The results of the research conducted are a classification for measuring the effectiveness of online learning in Pringsewu. The feasibility of the model obtained is supported by the results of the analysis of the Naïve Bayes model which has an accuracy rate of 98.48%, an AUC value of 0.995, a precision level of 98.17% and a 100% recall. In this study, the level of accuracy of the performance of the model used reached values above 90%. In addition, the AUC value of the two methods used is also more than 90% which is a value that is categorized as Excellent Classification. Further research can be carried out using other different parameters such as Economic Capability, Regional Location, Connectivity Mode, Digital Literacy, and others. In addition, this research was conducted only from the student's point of view. Inclusion of school opinion in future research will be useful in determining the exact effectiveness of online learning.