Decision Support in University Admissions Systems via the Use of Data Mining Techniques for Student Performance Prediction

Abstract

To choose students who will do well academically in universities, it is crucial to have an admissions system based on legitimate and trustworthy admissions criteria. Methods for employing data mining to aid colleges in admissions decisions are the subject of this research.
estimate how well prospective students will do in college. For the purpose of testing the validity of the suggested technique, we utilized a dataset consisting of 2,039 student records from a Computer Science and Information College at a Saudi public institution between 2016 and 2019. The findings show that certain pre-admission parameters (high school grade average, Scholastic Achievement Admission Test score, and General Aptitude Test score) may predict how well a student would do in their first year of college. The findings also suggest that a student's score on the Scholastic Achievement Admission Test is the best predictive factor for admission. Thus, this score has to be given additional weight in selection procedures. We also discovered that the Artifical Neural Network method outperforms the other classication methods (Decision Trees, Support Vector Machines, and Naive Bayes) with an accuracy rate of over 79%.