Comparison of SVM, NBC, and KNN Classification Methods in Determining Students’ Majors at SMK N02 Manokwari

Abstract

The stages of choosing a major for prospective SMK students are rarely the beginning of the next career determination. The determination of the major aims to make students more directed in receiving lessons based on the abilities and talents of the students, and, of course, when the student graduates, they already have the skills to get a job if they do not continue their education to college. The method used in this study is data mining techniques. But not all data mining algorithms perform well in classifying the selection of interest paths at the SMK level. Therefore, this study will discuss the comparative analysis of the performance level of the Support Vector Machine (SVM) classification algorithm and the Naïve Bayes Classifier (NBC) and K-Nearest Neighbors (KNN). Comparison of NBC, KNN and SVM methods was measured using feeding accuracy for the KNN method to get an accuracy of 54.56%, then for the NBC method to get an accuracy of 74.78%, and the SVM method to get an accuracy of 58.70%. Then it can be concluded that the three methods, based on the attributes used by the NBC method, got high accuracy, which is 74.78%.