A soft computing model for predicting the academic performance of first-year students

Abstract

This article examines a soft computing model designed to predict the academic success of first-year students under conditions of educational data uncertainty. The need to combine fuzzy logic, expert assessment, and intelligent analysis of educational data is substantiated. It is shown that academic performance prediction should be based on academic, motivational, behavioral, and social adaptation factors. The proposed model allows for the identification of risk groups, interpretation of the causes of performance decline, and support for management decisions in an educational institution.