Building an effective recommendation model for students' academic pathways using machine learning

Abstract EN

Selecting the most suitable academic pathway is critical for students in higher education, as it can significantly impact their future career opportunities and success.

this paper proposed a recommendation system for students in the civil department at al-Azhar university's faculty of engineering.

the system was created using the grades of students who graduated from the department in the period from 2018 to 2022.

different machine learning algorithms such as decision tree (DT), random forest (RF), k-nearest neighbor (KNN), naive bayes (NB), and support vector machine (SVM), are applied to create models for each major in the civil department.

the models are evaluated using 5-fold cross-validation method, with the F1-measure serving as the performance criterion.

the results indicate that the KNN model achieves the best performance for the structural engineering and public works majors, with an F1-measure of 84% and 87% accuracy, respectively.

on the other hand, the nb model achieves the best performance for irrigation and hydraulics, with an F1-measure of 69%.

the proposed recommendation system can potentially assist students in making informed decisions about their academic careers, by recommending the most appropriate major based on their academic performance.