A Comparative Analysis of Fuzzy Methods for Predicting Student Dropout Rate

Predicting student dropout accurately is a critical issue in the education system. This study employs fuzzy approaches - Analytic Hierarchy Process (AHP), Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), and Fuzzy Synthetic Evaluation - to forecast dropout rates. By calculating criteria weights, closeness coefficients, and fuzzy scores, we rank alternatives and identify key attributes influencing dropout rates. Our results show that admission grade, age at enrollment, previous qualification, and curricular units in the 1st semester are the most significant predictors, while scholarship holder status has relatively lower importance.