INTELLIGENT SYSTEM FOR CLASSIFICATION OF STUDENT PERSONALITY WITH NAIVE BAYES ALGORITHM

Abstract

Various kinds of problems arise from the changing personality behavior of students. Therefore, an intelligent system is needed to determine the personality type of students. This study applies an intelligent system with a classification method using the Naïve Bayes to determine the personality of the student based on Sanguine, Choleric, Melancholic, and Phlegmatic classes against Hippocrates-Galenus typology. The attributes used include gender, age, year of class, answers to test A, answers to test B, answers to test C, and answers to test D. System testing is carried out with a scheme for sharing training data and test data. The data used is questionnaire data based on the Hippocrates-Galenus typology which is filled out by 130 students. Then it is divided into training data to form a classification model of 117 data, and there are 13 pieces of data used as test data for accuracy testing. The proportion is 90:10 using 10-fold cross validation. The data held are then calculated using the nave Bayes algorithm. Based on the results, there were 12 students correctly predicted and 1 students did not predict correctly so that an accuracy of 92.31% was obtained with an error rate of 7.69%.