Proponent/Claimant
Eduardo Edu Cornillez
Abstract
The purpose of this study is to predict the performance of students in mathematics in the modern world using different machine learning algorithms. In this study, multiple linear regression, J48 decision tree algorithm, and Naïve Bayes classification algorithm were implemented to predict the performance of the students in mathematics in the modern world (MMW). This study employed a correlational and predictive research design where secondary data from the records of the teachers in MMW during the academic year 2019-2020 was used. The data are records of 144 education students which are composed of their ratings in attendance, quizzes, recitation, midterm exam, final exam, and their final grade. Results revealed that the majority of the students performed well in the subject having a very good rating. Also, the analysis revealed that ratings in attendance, quizzes, recitation, midterm exam, and final exam have a significant positive relationship with the final grades. Among the five variables, the most influential component to the overall performance of the students is the midterm exam rating. Using the 10-fold cross-validation, prediction models were also generated using Naïve Bayes and J48. Further, the Naïve Bayes algorithm provides better performance for predicting the students’ academic performance in MMW which provided a 73.61% accuracy, followed by the J48 (72.22%), and Multiple Linear Regression with 70.2% accuracy. Lastly, these machine learning models should be employed to improve the learning outcomes of students.
Name of Research Journal
TARAN-AWAN Journal of Educational Research and Technology Management
Volume and Issue No.
Volume 1, No. 1
Date/Year of Publication
2020
Citation
Morilla, R., Omabe, R., Tolibas, C. J., Cornillez Jr, E. E., & Treceñe, J. K. (2020). Application of machine learning algorithms in predicting the performance of students in mathematics in the modern world. TARAN-AWAN Journal of Educational Research and Technology Management, 1(1), 49-57.
