Stacked Ensemble Model for Predicting International University Admission

Abstract— The escalating demand for postgraduate education has heightened the competition for university admissions, prompting recent graduates to turn to expensive consultancy services due to their limited grasp of admission intricacies. This seminar introduces a sophisticated solution: a Stacked Ensemble machine learning approach that incorporates estimators including Logistic Regression, Support Vector Classifier (SVC), and Decision Trees, culminating in Bagging as the final estimator. This ensemble method [2][8] automates the precise prediction of postgraduate admission probabilities, empowering graduates to discern universities that best align with their profiles. A rigorous evaluation of these diverse strategies reveals Logistic Regression, SVC, and Decision Trees [8] as leading performers, offering highly accurate predictions. The proposal recommends the adoption of this ensemble approach to forecast the likelihood of prospective applicants securing admission to their preferred universities, providing an advanced solution for the benefit of aspiring postgraduate scholars.

Stacked Ensemble   Predicting International University Admission   Machine Learning  Admission Probability  Postgraduate Education