Implementation of Decision Tree and Naïve Bayes Classification Method for Predicting Study Period

Abstract

student learning performance influences the quality of a university. One indicator of assessment of student learning performance is the student's study period. By predicting the student's study period, universities can evaluate performance to strive to improve the quality of universities. Data mining is one of the choice technique in terms of predictions. But, each algorithm in data mining has the advantages of each, including Decision Tree and Naïve Bayes Algorithm which were tested in this research. This research shows Decision Tree is better than Naïve Bayes Algorithm to be the best choice to predicting study period with accuracy level tested for several ratios of training data and test data are 60:40,65:35,70:30,75:25 and 80:20 produce the highest accuracy values obtaines by decision tree with accuracy values are 90%, 89,14%, 89,3%, 88,8% and 88%. But, the amount of training data in this research does not affect the value of accuracy. This is shown from the value of accuracy in a certain ratio is very small but sometimes also the value of accuracy is high.