Evaluation of Academic Plans of Study Using Data Mining Techniques

• Published in: 2013 IEEE 13th International Conference on Advanced Learning Technologies pp. 224 - 228
• Main Authors: Siddiqui, Muazzam Ahmed, Gemalel-Din, Shehab
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.07.2013
• Subjects: Computational modeling; Data mining; Data models; educational data mining; Educational institutions; GPA prediction; plan of study assessment; predictive modeling; Predictive models; Regression tree analysis; Support vector machines
• ISSN: 2161-3761; 2161-377X
• DOI: 10.1109/ICALT.2013.70

A plan of study enumerates the courses recommended by an academic program along with a time frame to complete the requirements of a degree or credential. A general plan of study can be recommended by an academic department or college to all of its students or, as an alternative, personalized plans of study can be created for each student by his/her academic advisor. This paper presents a case study of assessing the recommended as well as personalized plans of study in terms of their affect on students' performance using data mining techniques. The study was conducted at the Information Systems Department, Faculty of Computing and Information Technology, King Abdulaziz University. We computed the degree to which each student followed the recommended plan of study and correlated this to the students GPA to assess the impact of the plan of study to the academic performance. Our results showed a statistically significant, moderate positive correlation indicating that following the recommended plan of study has a positive impact on the academic performance. To assess the personalized plans of study proposed by the academic advisor, we built and compared three models to predict the GPA resulting from these plans. Given a proposed plan of study, our model was able to predict the correct GPA with a 0.44 root mean squared error.