Evaluating the performance of ID3 method to analyze and predict students' performance in online platforms

Abstract

The primary goal of this work was to establish the influence of the epidemic on education, especially the impact of online platforms on students' overall performance. To improve education quality in this new normal, it is vital to ascertain the elements influencing students' performance. During the Covid-19 epidemic, online-based learning (e-learning) activities increased significantly as every educational institution shifted its operations to digital means. To improve education quality in this circumstance, it is vital to ascertain the elements that influence students' performance. Data mining techniques are becoming more used in educational field research. Educational data mining is a new area that tries to study the unique, ever more extensive data sets generated by educational settings to better understand the students who use them. In this study, Iterative Dichotomiser 3 was implemented to explore precisely 280 students' data to evaluate its performance on an online learning platform. 10-fold cross-validation and the percentage split method were used to assess the classifier. In this analysis, the 10-fold cross-validation method outperformed the percentage split by almost 3 percent, where this validation method achieved almost 77.86 percent accuracy. The effectiveness of the ID3 classifier in predicting students' performance on an online platform will be investigated in this research.