A Classification Model for Predicting the Suitable Study Track for School Students

Abstract

One of the most important issues to succeed in the academic life is to assign students to the right track when they arrive to the end of the basic education stage. The main problem in the selection of an academic track in basic Jordanian schools is the lack of useful knowledge for students to support their planning. This paper utilized data mining techniques to provide a classification approach to support basic school students in selecting the suitable track. For this purpose, a decision tree classification model was developed to determine which track is suitable for each student. There are a set classification rules that were extracted from the decision tree to predict and classify the class label for each student. A confusion matrix is built to evaluate the model where the 10-fold Cross Validation method was used for accuracy estimation of the model. The overall accuracy of the model was 87.9% where 218 students were correctly classified out of the 248 students.