A Machine Learning Approach for Graduate Admission Prediction

Abstract

With the increase in the number of graduates who wish to pursue their education, it becomes more challenging to get admission to the students' dream university. Newly graduate students usually are not knowledgeable of the requirements and the procedures of the postgraduate admission and might spent a considerable amount of money to get advice from consultancy organizations to help them identify their admission chances. However, giving the limited number of universities that can be considered by a human consultant, this approach might be bias and inaccurate. Thus, in this paper, a machine learning approach is developed to automatically predict the possibility of postgraduate admission to help graduates recognizing and targeting the universities which are best suitable for their profile. This paper evaluates three learning strategies of regression to predict the university rate given the students' profile; namely, linear regression, decision tree, and logistic regression model. This paper evaluates, these models to select the best model in terms of the highest accuracy rate and the least error. Logistic Regression model shows the most accurate prediction in our experiments, and hence, we suggest employing this model to predict the future applicant's university chance of admission.