To build a reliable classification model, the
CRISP-DM methodology (Cross-Industry
Standard Process for Data Mining) [8] is
adopted. The methodology consists mainly of
five steps: Collecting the relevant features of the
problem under study, preparing the data,building the classification model, evaluating the
model using one of the evaluation methods, and
finally using the model for future prediction of
the student performance. These steps are
presented in the next subsections.