Ensemble Learning Approaches to Strategically Shaping Learner Achievement in Thailand Higher Education

Thailand faces a severe problem of students dropping out of the higher education system. Therefore, this research has three critical objectives: (1) to study the context of students’ academic achievement in science and technology at the higher education level, (2) to assemble a model to predict the risk of students dropping out of higher education, and (3) to evaluate a model for predicting the risk of a student dropping out from higher education. The population and research sample were 2361 students’ academic achievements from five educational programs of the Faculty of Science and Technology at Rajabhat Maha Sarakham University during the 2010–2022 academic year. The research tool utilized data mining and supervised machine learning techniques: Decision Tree, Naïve Bayes, Neural Networks, Gradient Boosting, Random Forest, and Majority Voting. Model performance was evaluated using the cross-validation approaches and confusion matrix techniques, with four indicators: Accuracy, Precision, Recall, and F1-Score. The results showed that learners’ context in science and technology had various learning achievements. The educational program that needs to be monitored is the Bachelor of Science Program in Computer Science. This research successfully developed a predictive model for student dropout risk with an accuracy of 88.14% and an S.D. equal to 1.04. Therefore, this research dramatically benefits the public and stakeholders of Rajabhat Maha Sarakham University, who should be encouraged and encouraged to continue this research.