Designing and evaluating a big data analytics approach for predicting students’ success factors

Abstract

Reducing student attrition in tertiary education plays a significant role in the core mission and financial well-being of an educational institution. The availability of big data source from the Learning Management System (LMS) can be analysed to help with the attrition issues. This study aims to use an integrated Design Science Research (DSR) methodology to develop and evaluate a novel Big Data Analytical Solution (BDAS) as an educational decision support artefact. The BDAS as DSR artefact utilises Artificial Intelligence (AI) approaches to predict potential students at risk. Identifying students at risk helps to take timely intervention in the learning process to improve student academic progress for increasing their retention rate. To evaluate the performance of the predictive model, we compare the accuracy of the collection of representational AI algorithms in the literature. The study utilized an integrated DSR methodology founded on the similarities of DSR and design based research (DBR) to design and develop the proposed BDAS employing an specific evaluation framework that works on real data scenarios. The BDAS does not only aimto replace any existing practice but also support educators to implement a variety of pedagogical practices for improving students’ academic performance.