Endang Sugiharti, ILKOM UNNES and SAFIT FIRMANSYAH, ILKOM UNNES and FEROZA ROSALINA DEVI, ILKOM UNNES (2017) PREDICTIVE EVALUATION OF PERFORMANCE OF COMPUTER SCIENCE STUDENTS OF UNNES USING DATA MINING BASED ON NAÏVE BAYES CLASSIFIER (NBC) ALGORITHM. Journal of Theoretical and Applied Information Technology, 95 (4). pp. 902-911. ISSN 1992-8645

Preview	PDF Download (4MB) | Preview
Preview	PDF (PREDICTIVE EVALUATION OF PERFORMANCE OF COMPUTER SCIENCE STUDENTS OF UNNES USING DATA MINING BASED ON NAÏVE BAYES CLASSIFIER (NBC) ALGORITHM) - Published Version Download (2MB) | Preview

Abstract

Predictive evaluation is essential in order to map the performance of students from the Department of Computer Science of Mathematics and Natural Sciences Faculty of Universitas Negeri Semarang (UNNES), a state university in Semarang, Indonesia and graduation of students in a timely manner can also be predicted. This predictive evaluation can be seen by making a system based on the algorithm of Naïve Bayes Classifier (NBC). The data were taken from the performance of students which is the GPA from the 1st semester up to the 4th semester. The problem is how to predict the success of students in Computer Science Department of UNNES to graduate on time based on the performance of students from the 1st semester to the 4th semester? The main purpose of this research is to produce a system based on NBC algorithm that is able to predict the success of students to finish the study on time based on the performance of the students which is the GPA of the 1st semester to the 4th semester. To resolve this problem, the research divided into two stages of completion. The first stage was the literature review. This stage has been conducted by the researchers. The second stage determined the prediction of Computer Science student achievement using the method of NBC. This stage including (1) Data Collection, (2) Build a data mining system, (3) Data Processing, (4) Conducting the process of prediction, and (5) Analysis of Results. Based on the calculations of NBC that has been carried out, it can be concluded that 85% of students will graduate on time. The use of NBC will be better when more training data.

Actions (login required)

View Item