Improved fuzzy modelling to predict the academic performance of distance education students

Authors

  • Osman Yildiz Yildiz Technical University
  • Abdullah Bal Yildiz Technical University
  • Sevinc Gulsecen Istanbul University

DOI:

https://doi.org/10.19173/irrodl.v14i5.1595

Keywords:

Distance Education, academic performance, fuzzy logic, genetic algorithm, online learning,

Abstract

It is essential to predict distance education students’ year-end academic performance early during the course of the semester and to take precautions using such prediction-based information. This will, in particular, help enhance their academic performance and, therefore, improve the overall educational quality. The present study was on the development of a mathematical model intended to predict distance education students’ year-end academic performance using the first eight-week data on the learning management system. First, two fuzzy models were constructed, namely the classical fuzzy model and the expert fuzzy model, the latter being based on expert opinion. Afterwards, a gene-fuzzy model was developed optimizing membership functions through genetic algorithm. The data on distance education were collected through Moodle, an open source learning management system. The data were on a total of 218 students who enrolled in Basic Computer Sciences in 2012. The input data consisted of the following variables: When a student logged on to the system for the last time after the content of a lesson was uploaded, how often he/she logged on to the system, how long he/she stayed online in the last login, what score he/she got in the quiz taken in Week 4, and what score he/she got in the midterm exam taken in Week 8. A comparison was made among the predictions of the three models concerning the students’ year-end academic performance.

Author Biographies

Osman Yildiz, Yildiz Technical University

Informatics Department

Abdullah Bal, Yildiz Technical University

Electronics and Communications Engineering Department

Sevinc Gulsecen, Istanbul University

Informatics Department

Published

2013-12-10

How to Cite

Yildiz, O., Bal, A., & Gulsecen, S. (2013). Improved fuzzy modelling to predict the academic performance of distance education students. The International Review of Research in Open and Distributed Learning, 14(5). https://doi.org/10.19173/irrodl.v14i5.1595

Issue

Section

Research Articles