ALGORITMA NAIVE BAYES UNTUK MEMPREDIKSI JUMLAH SISWA BERPOTENSI DROP OUT

Abstract

The quality of education in schools can be seen from the high level of student success and low student failure. One indicator of student failure is the case of Drop Out). The Drop Out problem is something that is interesting to study, because it can affect the quality of education. In this study, the authors conducted research on the Naive Bayes Algorithm to predict the number of potential students to drop out (Case Study: SMK Negeri 2 Kotabumi, North Lampung). From the application of data mining, it can be seen that the data is processed using the Naive Bayes Algorithm and using Microsoft Excel. Based on the results of the application or prediction tools for drop out students using rapidminer 7.1, the results obtained from student score data taken at SMK Negeri 2 Kotabumi, North Lampung, there are 1178 student score data from 2017- 2020, it is known that on average 256 students do not drop outs and 34 drop outs per year with an average accuracy of 98.745% per year and an average classification error of 1.255% per year. And in testing 11 testing data, 10 students did not drop out and 1 student data was wrong with the status of not dropping out to drop out with an accuracy rate of 90.91% and classification error 9.09%.