In this research, we aimed at increasing college student retention by performing
early detection of academic risk using data mining methods.
Predicting students' academic performance is critical for educational institutions
because strategic programs can be planned in improving or maintaining students'
performance during their period of studies in the institutions. Data mining technologies
can be used to monitor students and simultaneously analyzing their academic behavior,
thus providing a basis for implementing necessary intervention procedures, if required.
The paper describes and lays out a methodological framework to develop model
that can be used to perform inferential queries on student performance using student
academic records.
Preliminary results on Academic Educational Data Mining (AEDM) model
development using Decision Tree data mining algorithms for classification are presented to
classify students as early as possible, into three groups: ‘low‐ risk' students who have a
high probability of success; ‘medium‐ risk' students who may succeed thanks to the
measures taken by the university; and the ‘high‐ risk' students who have a high probability
of failing where several classification rules were generated.