Predicting higher education student performance with educational data mining technique

Le résumé fourni par la source

Predicting student performance in higher education is critical for enhancing the academic outcomes of students. This study conducted on undergraduate students at STMIK Mikroskil and STIE Mikroskil in Medan, Indonesia, deals with a research gap by exploring factors beyond conventional metrics like cumulative grade point average (CGPA). The unique Indonesian educational landscape introduces additional factors, including graduation time and lecturer competency. Acknowledging the importance of student behaviour and lecturer competency, the study employs an artificial neural network model to predict student performance. By considering variables such as entry pathway, attendance, grade point average (GPA), scholarship, and lecturer performance index, the model achieves high accuracy - 85.33% for CGPA and 77.43% for graduation time. This research contributes to adopting educational data mining, aligning with Indonesian education regulations and facilitating early identification of at-risk students for targeted interventions.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.