Efficient Student Behaviour Analysis in E-Learning Using Data Mining Approaches

Abstract

Recognising and assessing how pupils act is essential for customising educational opportunities and enhancing educational results in online learning. In particular, Support Vector Machine (SVM), Decision Tree (DT), and Naive Bayes (NB) are employed in this work to analyse the characteristics of pupil conduct in online educational settings. The main goal is to determine the best strategy for thoroughly comprehending how students communicate in online learning environments. Employing metrics like RMSE (Root Mean Square Error), RSE (Relative Absolute Error), and RRSE (Relative Root Square Error) to evaluate the outcome of DM (Data Mining) methods. The results show that SVM regularly beats DT and NB throughout all criteria, showing that it has a greater capacity to identify complex relationships in pupil activity records with RMSE of 0.02714, RAE of 0.00279 and RRSE of 0.02117, respectively. The tool used for execution is Jupyter Notebook, and the language used is Python.