Prediksi Dampak Pembelajaran Hybrid Learning Menggunakan Naive Bayes

Abstract

This research use to predict the impact of hybrid learning on Medan State Polytechnic students. This algorithm was chosen because it has excellent performance in classification compared to other algorithms. Statistical and probabilistic methods are used in the operation of this algorithm to make predictions about what will happen in the future. Technology mastery, level of teacher-student interaction, and mastery of teaching materials are the variables used in this study. The sample data used came from students of the Software Engineering Technology Study Program of Medan State Polytechnic. The prediction results carried out manually with naïve bayes, with training data of 100 (one hundred) students and test data of 1 (one) student, produced a result of 0.012, which indicates an increase in student academic results. The test results were proven using the phyton programming language. The first test results, with 20% test data, resulted in an increase in academic results by 86% around 13 students with an accuracy value of 80%, and the second test, with 40% test data, resulted in an increase in academic results by 92% around 29 students with an accuracy value of 88%.