Classification of Productive Subject Value with Vector Machine Support Method

Abstract

This research aims to determine student competency obtained from productive subjects, one of the subjects that has an important role in measuring student quality. With the development of technology, student evaluations of productive subject grades can use data mining methods, one of which uses the Support Vector Machine approach. This method aims to build a classification of students' productive subject scores by identifying the variables that influence them. Based on the research results, it can be concluded that the value of productive subjects at Public Vocation High School 1 Pasuruan can be predicted and evaluated using data mining techniques that utilize the Support Vector Machine algorithm. It aims to predict student grades in productive subjects by utilizing a logistic regression algorithm. The classification results illustrate that student grades in productive subjects are influenced by learning media variables, teacher teaching quality, classroom/laboratory facilities, interest, and motivation to learn. Apart from that, the Support Vector Machine algorithm has an accuracy value of 64.5001, precision of 56.0279, recall = 86.9952, and F1-Score = 67.882.