A MULTI-PARAMETER INTELLIGENT MODEL FOR EARLY PREDICTION OF STUDENT EXPULSION RISK IN HIGHER EDUCATION INSTITUTIONS (BASED ON THE EXPERIENCE OF THE CREDIT-MODULE SYSTEM)
Abdurasulov Olqar Toir o‘g‘li
Registrar’s Office, Samarkand Branch of ISFT Institute Chief Specialist in Databases
Temirov Anvar Amirovich
Vice-Rector for Academic Affairs, ISFT Institute PhD in Pedagogical Sciences, Associate Professor
Keywords: credit module system, student retention, dropout risk, early warning system, multifactor analysis, attendance monitoring, learning analytics.
Abstract
This article describes a method for identifying the risk of student dismissal early in the semester at higher education institutions. The aim of the study was to develop a simple and practical model that assesses risk using five indicators under the credit module system and to test how well it works. The model combines class absences, delays in tuition payments, midterm results, activity on the online learning platform and failure to answer calls into one 100 point scale. Students are placed into green, yellow and red zones according to their risk level. The model was tested at the Samarkand branch of the ISFT Institute in the 2025/2026 academic year with about 7,000 students and compared with the results of 2024/2025. The share of students dismissed at the end of the semester fell from 7.8% to 1.9%, and overall retention rose from 85.2% to 97.4%. The results suggest that the model can be used at other universities, although further testing is needed.
