Hybrid Text Ensemble-MultiEmbed: A Novel Machine Learning Approach for Automated Learning Outcome Prediction

Abstract

As technology continues to shape education, the need for effective tools to predict learning outcomes and tailor educational experiences becomes paramount. This research paper presents a model for the automatic prediction of learning outcomes, leveraging ensemble learning techniques, namely bagging, boosting and voting. These ensemble methods are explored for their ability to enhance predictive accuracy and robustness in educational analytics. This research intended to evaluate the academic performance of students and build an efficient classification model through the fusion of single and ensemble-based classifiers for the automated mapping and classification of learning outcomes. Five single classifiers Naive Bayes (NB), Random Forest (RF), Decision Tree (DT), Support Vector Classifier (SVC), Logistic Regression (LR) were observed along with three well established ensemble algorithms encompassing Bagging (BAG), AdaBoost (MB), and Voting (VT) independently. The proposed model Hybrid Text Ensemble-MultiEmbed (HTME) outperformed well with mean F1-score of 96.25% for all four learning outcomes. The study implies that the proposed model could be useful in evaluate the academic performance by mapping and predicting the learning outcomes.