Track: Data Analytics

Abstract

Increasing student admission and enrollment, especially in engineering and computing programs, is a desirable goal for many universities. At the same time, this goal can be difficult to achieve. The aim of this research is to develop a data analytics model that can be used by universities and colleges to improve student admission and enrollment process. Predictive analytics is the technique of using historical data to create, test and validate a model to best describe and predict the probability of an outcome. In recent years, predictive analytics has been used in many areas including manufacturing, healthcare, and service industry. In engineering and computer science education, data analytics models can be used to describe and predict what will happen during the different stages of the enrollment process. This can help an institution determine the interventions that should be taken to support students or meet recruiting goals. In this innovative practice paper, we develop analytics models based on logistic regression, neural networks, and decision trees utilizing historical data from a local university. We focus on the analysis and modeling of student admission and enrollment data to provide a decision support for the admission staff. It may be noted, however, that this model cannot be stand alone and only serves to compliment university administrators' decision-making process to manage admissions and enrollments effectively. The developed models are tested and validated using k-fold cross validation technique.