FRAMEWORK FOR IDENTIFYING THE MOST LIKELY SUCCESSFUL UNDERPRIVILEGED TERTIARY STUDY BURSARY APPLICANTS

Renier Steynberg, Danie Lötter, Jan Harm van Vuuren

Abstract


In this paper, a decision support system framework is proposed that may be used to assist a tertiary bursary provider during the process of allocating bursaries to prospective students. The system identifies those in an initial pool of applicants who are expected to be successful tertiary students, to facilitate final selection from a shortlist of candidates. The working of the system is based on various classification models for predicting whether bursary applicants will be successful in their respective tertiary studies. These model predictions are then combined in a weighted fashion to produce a final prediction for each student. In addition, a multi-criteria decision analysis method is used to assign each of the applicants to a ranking level. In this way, the system suggests both a predicted outcome for each candidate and a ranking according to which candidates may be compared. The practical working of the system is demonstrated in the context of real data provided by an industry partner, and the success rate of the system’s recommendations is compared with that of the industry partner.


Keywords


Decision support; statistical learning; classification; ensemble models; multi-criteria decision analysis; tertiary study bursaries

Full Text:

PDF


DOI: http://dx.doi.org/10.7166/28-2-1695

Refbacks

  • There are currently no refbacks.




Copyright (c) 2017 Renier Steynberg, Danie Lötter, Jan Harm van Vuuren


ISSN 2224-7890 (on-line) ; ISSN 1012-277X (print)


Powered by OJS and hosted by Stellenbosch University Library and Information Service since 2011.


Disclaimer:

This journal is hosted by the SU LIS on request of the journal owner/editor. The SU LIS takes no responsibility for the content published within this journal, and disclaim all liability arising out of the use of or inability to use the information contained herein. We assume no responsibility, and shall not be liable for any breaches of agreement with other publishers/hosts.

SUNJournals Help