Performance Based Prediction of the Students in the Physics Subject using Traditional and Machine Learning Approach at Higher Education Level
Abstract
In higher educational institutions, it is not an easy task to judge the performance of the students timely which is becoming more challenging. Although institutions have gathered a lot of data about their students. They do not have some specific methods to extract meanings from it. The main objective of this study was to find out the performance-based prediction of the students using their demographic and academic factors by using traditional and machine learning approaches. Graduates and undergraduate students studying in KUST were the population of the study. The study was delimited to the department of physics. A total of ninety graduate and undergraduate students were selected randomly using a simple random sampling technique as the entire sample. The result indicated that percentage in matric (Correlation = 0.304), intermediate (Correlation = 0.245) and National Aptitude Test scores (Correlation = 0.480) found the best predictors. Further research was recommended to predict students’ academic performance by taking other aspects of the students like personality, cognitive, psychological, and economic domain for making a dataset of the features which may be used in machine learning approach which is more reliable to judge the academic performance of the students at the higher education level.
Keywords: Performance, Challenging, Demographic, Prediction, Examination
Copyright (c) 2020 International Journal of Innovation in Teaching and Learning (IJITL)

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
IJITL by Department of Education, IIUI is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. Based on a work at www.iiu.edu.pk. Permissions beyond the scope of this license may be available at Licensing and Copyright. All articles published by IJITL are licensed under the Creative Commons Attribution 4.0 International License. This permits anyone to copy, redistribute, transmit and adapt the work provided the original work and source is appropriately cited as specified by the Creative Commons Attribution License.The journal allows readers to freely read, download, copy, distribute, print, search, or link to the full texts of its articles and to use them for any other lawful purpose. Once published the copyrights are retained with the Journal.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International