Student Engagement Level Prediction in Online Education

This paper presented a novel framework for predicting student engagement levels in online education environments using an AHP-guided Stacking Ensemble model with an MLP meta-learner. By combining principled feature weighting through AHP with the complementary predictive strengths of multiple base classifiers, the proposed approach achieves superior accuracy, interpretability, and scalability compared to conventional methods. The experimental evaluation demonstrated that the proposed system achieves 91.2% accuracy and a ROC-AUC of 0.96, substantially outperforming Logistic Regression, Decision Trees, Random Forest, and standard MLP baselines. The integration of parent engagement features alongside behavioral indicators further enriched the predictive model, capturing holistic dimensions of student engagement. In future work, the system can be extended by incorporating real-time LMS log data streams for continuous engagement monitoring. Deep learning architectures such as Long Short-Term Memory (LSTM) networks could be explored to model temporal engagement patterns over the course of a semester. Additionally, natural language processing techniques can be applied to analyze discussion forum posts and assignment submissions to capture cognitive engagement dimensions. Expanding the framework to support multilingual and multicultural learner populations will further enhance its global applicability.

[1]    P. Kundur et al., "Definition and classification of power system stability," IEEE Trans. Power Syst., vol. 19, no. 3, pp. 1387–1401, Aug. 2004. [2]    Fredricks, J. A., Blumenfeld, P. C., & Paris, A. H. (2004). School engagement: Potential of the concept, state of the evidence. Review of Educational Research, 74(1), 59–109. [3]    Deng, R., Benckendorff, P., & Gannaway, D. (2019). Progress and new directions for teaching and learning in MOOCs. Computers & Education, 129, 48–60. [4]    Almosallam, E. A., & Ouertani, H. C. (2019). Learning Analytics: Definitions, Applications and Related Fields. In Advances in Data Science and Information Engineering (pp. 300–313). Springer. [5]    Chow, A. F., & Mullan, J. A. (2010). Faculty satisfaction and continuing challenges with course management systems. International Journal on E-Learning, 9(2), 175–190. [6]    Connell, J. P., Spencer, M. B., & Aber, J. L. (1994). Educational risk and resilience in African-American youth. Child Development, 65(2), 493–506. [7]    Wang, M. T., & Fredricks, J. A. (2014). The reciprocal links between school engagement, youth problem behaviors, and school dropout during adolescence. Child Development, 85(2), 722–737. [8]    Saaty, T. L. (1980). The Analytic Hierarchy Process. McGraw-Hill, New York. [9]    Chen, G., Davis, D., Lin, J., Hauff, C., & Houben, G. J. (2016). Predicting MOOC dropout over weeks using machine learning methods. In EMNLP Workshop on Machine Learning for Education. [10] Haykin, S. (1998). Neural Networks: A Comprehensive Foundation (2nd ed.). Prentice Hall.