International Journal of Information and Communication Technology Research

International Journal of Information and Communication Technology Research>> Call for Papers(CFP)>>Volume 7, Number 9, September 2017

International Journal of Information and Communication Technology Research


Prediction the Loyal Student Using Decision Tree Algorithms

Full Text Pdf Pdf
Author(s) Saeide kakavand, Taha Mokfi, Mohammad Jafar Tarokh
On Pages 32-37
Volume No. 4
Issue No. 1
Issue Date January, 2014
Publishing Date January, 2014
Keywords component; data mining, decision tree algorithms, loyal student, higher education, studentís performance


Abstract


One of the most important challenges that higher education system facing today is Providing more effective, efficient and higher quality education service to students, and predicting the pattern of loyal students. Because the universities are trying to raise educational quality, Applying data mining in higher education helps the manager, lecturer, and students to make higher performance. The aim of research paper is to understand the external factors that may cause the student loyalty. By doing that, the university can identify students who have decided to continue studying, so it can invest on them, and thus increase its educational quality. One of the best ways to achieve this is by using valid management and processing of the students database. In this study, using dataset from the Private University and applying data mining techniques, classify master students based on input characteristics and finally the pattern of faithful students (students who have decided to continue studying)were extracted. Classified students are based on personal information of students, student academic status, type of their pervious university (private or state university), finances and occupation status, and educational status of their parents. To classify students, the rule generation process is based on the decision tree algorithms like C.5, CART and CHAID. The results showed that CART decision tree algorithm is the best predictor with 94% accuracy on evaluation sample.

Seperator
    Journal of Information and Communication Technology | Journal of Science and Technology     
Copyrights
© 2014 IRPN Publishers