A data mining approach for investigating students' completion rates

One of the major challenges faced by higher education institutions is to enhance the quality of decisions made from knowledge derived from rapidly growing educational data. Data mining techniques are investigative tools that are used to extract significant unknown information from large data sets. This paper proposes to discover the most appropriate data mining technique(s) to investigate the relationship between prior learning, temporal sequence of courses and student performance attributes namely GPA and time-to-degree (number of semesters taken towards graduation) and later the correlation between GPA and time-to-degree. Once the relationships are established, it is proposed to find the optimized sequence of courses taken by successful students from similar prior learning backgrounds that would facilitate current/future students to graduate on time with high scores. More specifically this paper highlights this research gap from the literature review which will be further analysed by the authors using data mining.