Predictive analytics of university student intake using supervised methods

• Published in: IAES International Journal of Artificial Intelligence
• Main Author: Basheer M.Y.I.; Mutalib S.; Hamid N.H.A.; Abdul-Rahman S.; Malik A.M.A.
• Format: Article
• Language: English
• Published: Institute of Advanced Engineering and Science 2019
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85079484679&doi=10.11591%2fijai.v8.i4.pp367-374&partnerID=40&md5=4ae63e24abc095304a45926e20db4dfb

Predictive analytics extract important factors and patterns from historical data to predict future outcomes. This paper presents predictive analytics of university student intake using supervised methods. Every year, universities face a lot of academic offer rejection by the applicants. Hence, this research aims to predict student acceptance and rejection towards academic offer given by a university using supervised methods subject to past student intake data. To solve this problem, a lot of past studies had been reviewed starting from nineties era till now. From the analysis, two algorithms had been selected namely Decision Tree and k Nearest Neighbor. The dataset of past student intake was obtained with fifteen attributes, which are applicants’ gender, applicants studied stream during Sijil Peperiksaan Malaysia (SPM), university campuses, applicants’ hometown, disability, campus visit, course choice order in application form, applicant’s six SPM subjects result, orphan and status of acceptance. Several experiments were implemented to find the best model to predict the student’s offer acceptance by evaluating the model accuracy. Both models yield best accuracy at 66 percent with the selected attributes. This research gives a huge impact in selecting which applicants is suitable to be offered as well as adapting the university’s academic offering process in much intelligence way in the future. © 2019 Institute of Advanced Engineering and Science. All rights reserved.