<![CDATA[One issue that higher education institutions face is dropout rates. The issue of students quitting or dropping out of school can affect a college's reputation, accreditation, and long-term viability. The institution, the individual student, their family, and human resource development are all negatively impacted by student dropout. Universities must thus predict the probability of student dropout to take early preventive action. This study focuses on developing a prediction model for dividing students into three groups: successful dropouts, enrolled, and graduates, using data processing and machine learning approaches. The research data will include academic grades, semesters, attendance, economic background, occupation, marital or single status, and e-learning. High research results using these parameters and the SVM algorithm achieved an accuracy of 78.5%. A degree of accuracy was successfully achieved by the research results using these parameters and the Support Vector Machine (SVM) method, suggesting that the Support Vector Machine (SVM) algorithm is appropriate for predicting student dropout in higher education.]]>
Copyrights © 2026
