Fuzzy logic model for the selection of applicants to university study programs according to enrollment profile

Nemias Saboya, Omar L. Loaiza, Juan J. Soria, Jose Bustamante

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

8 Scopus citations

Abstract

This study consisted in elaborating a diffuse model that allows for the selection of applicants to university study programs considering their enrollment profile, based on the need to reduce errors of precision in the selection procedures and to minimize the human limitations of the evaluators who are faced with a wide variety of differentiated evaluation criteria for the study programs. This fuzzy model is composed of three input linguistic variables referring to knowledge competencies and an output variable regarding the enrollment profile level. It is also made up of 125 fuzzy rules developed by the Mamdani and centroid methods and validated by experts. The results of the application show its effectiveness with p < 0.05. The model is flexible and useful to ideally select as many new entrants as possible, and thus it offers advantages to comply with any institutional regulatory context and external quality assurance entities such as the SINEACE or others.

Original languageEnglish
Title of host publicationDigital Science
EditorsTatiana Antipova, Alvaro Rocha
PublisherSpringer Verlag
Pages121-133
Number of pages13
ISBN (Print)9783030023508
DOIs
StatePublished - 2019
EventInternational Conference on Digital Science, DSIC 2018 - Budva, Montenegro
Duration: 19 Oct 201821 Oct 2018

Publication series

NameAdvances in Intelligent Systems and Computing
Volume850
ISSN (Print)2194-5357

Conference

ConferenceInternational Conference on Digital Science, DSIC 2018
Country/TerritoryMontenegro
CityBudva
Period19/10/1821/10/18

Fingerprint

Dive into the research topics of 'Fuzzy logic model for the selection of applicants to university study programs according to enrollment profile'. Together they form a unique fingerprint.

Cite this