Performance in academic courses is among the most important factors affecting the quality of higher education available to students. In this paper, we use data mining techniques, specifically classification, to analyze students’ grades in different evaluative assignments for a course on data structures. For this purpose, we compare three different classifiers using real data from King Saud University to predict students’ performances. We apply classification techniques to both numerical and categorized attributes. Our results show that the model based on the Naïve Bayes algorithm provides the most accurate predictions. In addition we were able to obtain a model with a 91% accuracy to predict students failure in the course. © 2005 - 2015 JATIT & LLS. All rights reserved.
Mendeley saves you time finding and organizing research
Choose a citation style from the tabs below