Identification of Best Performing Students Using Machine Learning Algorithms: A Case Study of the Electrician Training Programme at Vaal University of Technology

The Vaal University of Technology (VUT) has made commendable progress in bridging the skills gap in electrical engineering in South Africa through its French-South African Schneider Electric Education Center (FSASEC). This initiative focuses on training young, underprivileged students to become proficient occupational electricians. Given the critical need for skilled artisans, the project synergises efforts from industry, academia, and private and public sectors. In seeking to enhance the selection process for its training programme, VUT employed machine learning techniques to refine its admission criteria, taking into account students’ performance in English, mathematics, and physical science. This approach stemmed from identifying a robust correlation between these subjects and the students’ subsequent success in the programme. The study achieved a 100% prediction accuracy rate using multiple regression analysis, effectively distinguishing potential high performers from those at risk of underperformance. This research underscores the effectiveness of integrating machine learning tools in educational admissions processes, ensuring that training programmes meet the immediate skills demands and equip students with essential technical and soft skills for substantial contributions to the workforce.