Employing Various Data Mining Techniques to Forecast the Success Rate of Information Technology Education Students

Abstract

This study was designed to investigate the factors that affect the success rate of Information Technology Education students which composed of Computer Science and Information Technology. Several variables such as years of graduation, entrance exams, and other variables have been used for the investigation. Several data mining techniques such as linear regression, neural network, and decision tree have employed to determine the valid predictors and acceptability of the data mining technique. The results show that the best predictor taken from the entrance exams is non-verbal ability while the best forecasting using data mining is decision tree analysis with 99.19 percent accuracy. If the results taken from the system will be incorporated in entrance examinations results, admission office will be able to identify students that can graduate on-time and whose students should be taken as probationary in the program. It can also identify students not to be taken in the program to avoid waste of time in studying at the University.