Mathematical model design to integrate EDM with KDDM framework

Higher Education institutions (HEIs), which face many challenges related to student performance, can benefit from integrating educational data mining (EDM) with knowledge discovery and data mining (KDDM). Through this integration, HEIs can harness the full potential of data to enhance the learning experience, improve educational outcomes, and drive innovation in teaching and learning. Discovering course-taking patterns in educational datasets in conjunction with contextual information has been a particularly challenging area. While this research contended that current KDDM processes fail to generate patterns associated with contextual information, this research attempted to make that case. This research paper began by establishing a relationship between cumulative grade point average (CGPA), time to degree, student course-taking patterns, and contextual factors (such as course difficulty patterns and number of courses). The motivation of this paper is to find out if EDM is capable of discovering this relationship and if not, will the integration of EDM with KDDM help in this regard. The contribution of this paper is that it has developed the relationship between the variables CGPA, time to degree, course taking pattern of students and course difficulty pattern. It has also contributed by conducting experiments on EDM thereby concluding that it cannot help much without KDDM and then by using KDDM (CRISP-DM) model with the integration of EDM to test this relationship. Furthermore, it was found that although this integration partially helped in establishing this relationship, it could not discover the contextual factors.