An Effective Ensemble Model to Predict Employment Status of Graduates in Higher Educational Institutions

Abstract

Several higher education institutions face the issue or difficulty of graduating more than 90% of their students who can competently satisfy and meet the industry's requirements. However, the industry is also challenged by the difficulty of locating skilled tertiary institution graduates who meet their requirements. The success or failure of any organisation is primarily determined by how its workforce is recruited and retained. As a result, one of the major and critical problems of management decision-making is the selection of an acceptable or satisfactory candidate for the job position. As a result, this work proposes a modern, accurate, and worthy machine learning classification model that can be deployed, implemented, and used to make predictions and assessments on job applicant attributes from academic performance datasets to meet the industry's selection criteria. This study took into account both supervised and unsupervised machine learning classifiers. Naive Bayes, MLP, Simple Logistic, Adaboost, Bagging and Ensemble Model are chosen for analysis. The proposed model outperforms other reported methods with an accuracy of 98.4253%.