Authors: Shajeea Khan; Jacintha Menezes
Addresses: Majan College, University College, Muscat, Oman ' Majan College, University College, Muscat, Oman
Abstract: Students' academic performance and progression is the key performance indicator of any educational institution. Students' academic progression may vary due to various factors. Therefore, identifying students at risk at an early stage could help to improve the academic progression of students, through academic counselling and close monitoring. Predictive analytics can discover patterns in data that can lead to meaningful predictions for the unknown data. In the proposed research, students already identified at risk are further categorised into low risk, medium risk and high risk. Student study level, type of assessment, type of module and issues faced by the students are taken as the influencing factors. The research focused on finding the influencing factors with the level of risk through predictive modelling. Research results show the issues faced by the student (academic/personal) as the main influencing factor with type of risk.
Keywords: data mining; knowledge discovery; WEKA; predictive analytics; classification; decision tree.
International Journal of Technology Transfer and Commercialisation, 2020 Vol.17 No.1, pp.68 - 75
Accepted: 13 Jun 2019
Published online: 15 Apr 2020 *