PROJECT TITLE :

Usage of Machine Learning for Strategic Decision Making at Higher Educational Institutions

ABSTRACT:

Higher Educational Institutions (HEIs) make policies, plans, and actions that are influenced by strategic decisions made at the institutional level. This article depicts decision-making systems at HEIs and their usefulness in promoting institutional governance. Because of the stakeholders' disengagement and the lack of effective computational methods, the decision-making process takes longer; 2) the "full picture" is not included, together with all relevant facts; and 3) the decision has a modest academic influence, among other things. Machine learning is an emerging subject of artificial intelligence that analyzes data and provides a greater knowledge of the material included in a certain context using various methods. We focus on assisting strategic decision-making, with deans' concerns being the most important mission to support, based on the author's prior efforts. Three supervised classification techniques are used in this research to estimate graduation rates using real data about South American undergraduate engineering students. To compare and evaluate Decision Tree, logistic regression, and Random Forest, the analysis of receiver operating characteristic (ROC) curve and accuracy is used as a measure of efficacy, with the latter showing the best results.