Machine Learning Approach to Determining the Influence of Family Background Factors on Students’ A
Authors:
OKOLIE Samuel
Publication Type: Journal article
Journal:
ISSN Number:
0
Downloads
18
Views
Abstract
Machine learning has been successfully applied to many domains such as
fraud detection, medicine, banking, bioinformatics, and so on. The application
of this paradigm to enhancing and evaluating the higher education tasks is a
new research area. There have been various works bordering on students’
performance, and related problems but the focus of this work is on applying
machine learning algorithms to students’ data for predictive purposes in an
educational environment. We propose to develop novel approach based on
machine learning algorithms to be delivered to educational institutions for
guiding their planning of educational activities within the scope of increasing
the academic performance of the students by taking into account their family
background factors. One thousand five hundred (1500) records of students in
three Nigerian tertiary institutions will be used. The students’ academic
performance will be measured by Cumulative Grade Point Average (CGPA) at
the end of first year. Waikato Environment for Knowledge Analysis (WEKA)
and See5 will be used to generate three decision tree models, Artificial Neural
Networks and two rulesets. These algorithms will be compared based on their
accuracy level and confusion matrices to determine the optimal model. The
rules generated from the optimal model will be disseminated to educational
administrators to guide in their planning activities.