Extracting Meaningful Features and Hidden Patterns from the Students Records Using Data Mining Techniques

Abstract

Educational Data Mining (EDM) helps to recognise the performance of students and predict their academic achievements that include the successes and failures aspects, also negative and challenges aspects. Consequently, a massive amount of students' data in educational systems has been collected, which has become difficult for officials to search through and obtain the knowledge required to discover challenges facing students and universities by traditional methods and this is a time-consuming task. The main aim of this research is to extract hidden patterns in students' historical academic data. The data mining tools used are classification and regression models that predict performance based on grade point average (GPA) and extraction of the frequent patterns generated by the association rules. The research data sets gathered from the College of Business and Economics (CBE) at Qassim University in KSA from 2014 to 2018. Knowledge from this can help in making appropriate decisions for certain circumstances and suggestions for overcoming students' weaknesses and failures. The findings show numerous problems related to a student's performance at different levels and in various courses. The results of the association rules indicated that there was a link between general education courses and student success. It was concluded that GPA failure is closely related to the first levels of study. The research outcomes indicated that there are many significant problems. Consequently, this study suggested a set of suitable solutions, which can be presented to the College of Business and Economics for the benefit and improving student performance and activating academic advising, in addition to enhancing the educational process.