Massive open online courses have revolutionised the learning environment, but their effectiveness is undermined by low completion rates. Traditional dropout prediction models in MOOCs often overlook complex factors like temporal dependencies and context-specific variables. These models are not adaptive enough to manage the dynamic nature of MOOC learning environments, resulting in inaccurate predictions and ineffective interventions. Accordingly, MOOCs dropout prediction models require more sophisticated artificial intelligence models that can address these limitations. Moreover, incorporating feature selection methods and explainable AI techniques can enhance the interpretability of these models, making them more actionable for educators and course designers. This paper provides a comprehensive review of various MOOCs dropout prediction methodologies, focusing on their strategies and research gaps. It highlights the growing MOOC environment and the potential for technology-driven gains in outcome accuracy. This review also discusses the use of advanced models based on machine learning, deep learning, and meta-heuristics approaches to improve course completion rates, optimise learning outcomes, and provide personalised educational experiences.
High dropout rates remain a significant challenge in Massive Open Online Courses (MOOCs), making early identification of at-risk students crucial. This study introduces a novel approach called In-Session Stacked Ensemble Learning for Dropout Prediction (ISELDP), which predicts student dropout during course sessions by combining multiple base learners—Adaptive Boosting (AdaBoost), Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Gradient Boosting—into a stacked ensemble with a Multi-Layer Perceptron (MLP) serving as the meta-learner. To optimise model performance, hyperparameters were tuned using Grid Search. The proposed method was evaluated under two scenarios using in-session student interaction data, one with imbalanced data and another with balanced data. Results demonstrate that ISELDP achieves an average accuracy of 88%, outperforming individual baseline models with improvements of up to 2% in accuracy and 2.4% in F1-score.
High dropout rates on in-session learning platforms pose a significant challenge to student retention and the overall success of educational programmes. This study proposes a novel framework that integrates multi-level stacked ensemble learning with optimised feature selection using a hybrid approach combining Genetic Algorithm (GA) with Correlation-Based Feature Selection (CFS). The model employs a Multi-Layer Perceptron (MLP) as a meta-learner, aggregating predictions from multiple ensemble-based base classifiers to enhance predictive accuracy. To improve generalisation and reduce noise, the proposed approach applies GA-CFS-driven feature optimisation in conjunction with data balancing techniques. Experimental results demonstrate that the proposed model outperforms benchmark approaches, achieving improvements of up to 22% in prediction accuracy and 12% in F1-score over standard stacked ensemble methods. These results highlight the effectiveness of combining meta-heuristic optimisation with ensemble learning to advance dropout prediction in online learning environments.
15 December 2023 in Proceedings on Engineering Sciences