Comparative Analysis of Self-Labeled Algorithms for Predicting MOOC Dropout: A Case Study
Source abstract
Abstract
Massive Open Online Courses (MOOCs) have expanded global access to education but continue to struggle with high attrition rates. This study presents a comparative analysis of self-labeled Semi-Supervised Learning (SSL) algorithms for predicting learner dropout. Unlike traditional supervised models that rely solely on labeled data, self-labeled methods iteratively exploit both labeled and unlabeled instances, alleviating the scarcity of annotations in large-scale educational datasets. Using real-world MOOC data, ten self-labeled algorithms, including self-training, co-training, and tri-training variants, were evaluated across multiple labeled ratios. The experimental results show that ensemble-based methods, such as Co-training Random Forest, Co-Training by Committee, and Relevant Random subspace co-training, achieve predictive accuracy comparable to that fully supervised baselines even with as little as 4% labeled data. Beyond predictive performance, the findings highlight the scalability and cost-effectiveness of self-labeled SSL as a data-driven approach for enhancing learner retention in massive online learning.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Online Learning and Analytics · Intelligent Tutoring Systems and Adaptive Learning · Educational Technology and Assessment
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Mathematics, University of Patras, 26504 Patras, Greece
School of Social Sciences, Hellenic Open University, 26331 Patras, Greece
Department of Management Science and Technology, University of Patras, 26334 Patras, Greece
School of Science and Technology, Hellenic Open University, 26335 Patras, Greece
References (51)
Show source-provided references
1. Yuan, L., and Powell, S. (2013). MOOCs and open education: Implications for Higher Education: A white paper. JISC Cetis, 1–21.
2. Feng, W., Tang, J., and Liu, T.X. (February, January 27). Understanding dropouts in MOOCs. Proceedings of the AAAI Conference on Artificial Intelligence, Honolulu, HI, USA.
3. Kizilcec, R.F., Piech, C., and Schneider, E. (2013, January 8–12). Deconstructing disengagement: Analyzing learner subpopulations in massive open online courses. Proceedings of the Third International Conference on Learning Analytics and Knowledge, Leuven, Belgium.10.1145/2460296.2460330 ↗
4. MOOCs: A systematic study of the published literature 2008–2012
5. A learning analytics methodology for understanding social interactions in MOOCs
6. Viteli, J., and Leikomaa, M. (2014, January 23–26). MOOCs completion rates and possible methods to improve retention-A literature review. Proceedings of the EdMedia 2014–World Conference on Educational Media and Technology, Tampere, Finland.
7. A systematic review for MOOC dropout prediction from the perspective of machine learning
8. Kloft, M., Stiehler, F., Zheng, Z., and Pinkwart, N. (2014, January 25–29). Predicting MOOC dropout over weeks using machine learning methods. Proceedings of the EMNLP 2014 Workshop on Analysis of Large Scale Social Interaction in MOOCs, Doha, Qatar.10.3115/v1/W14-4111 ↗
9. Zhu, X., and Goldberg, A. (2009). Introduction to Semi-Supervised Learning, Morgan & Claypool Publishers.10.1007/978-3-031-01548-9 ↗
10. On the effectiveness of self-training in MOOC dropout prediction10.1515/comp-2020-0153 ↗
11. Dalipi, F., Imran, A.S., and Kastrati, Z. (2018, January 18–20). MOOC dropout prediction using machine learning techniques: Review and research challenges. Proceedings of the 2018 IEEE Global Engineering Education Conference (EDUCON), Canary Islands, Spain.10.1109/EDUCON.2018.8363340 ↗
12. Kostopoulos, G., Kotsiantis, S., and Pintelas, P. (2015, January 1–3). Estimating student dropout in distance higher education using semi-supervised techniques. Proceedings of the 19th Panhellenic Conference on Informatics, Athens, Greece.10.1145/2801948.2802013 ↗
13. Li, W., Gao, M., Li, H., Xiong, Q., Wen, J., and Wu, Z. (2016, January 24–29). Dropout prediction in MOOCs using behavior features and multi-view semi-supervised learning. Proceedings of the 2016 International Joint Conference on Neural Networks (IJCNN), Vancouver, BC, Canada.10.1109/IJCNN.2016.7727598 ↗
14. Zhou, S., Cao, L., Zhang, R., and Sun, G. (2022, January 19–22). M-ISFCM: A semisupervised method for anomaly detection of MOOC learning behavior. Proceedings of the International Conference of Pioneering Computer Scientists, Engineers and Educators, Chengdu, China.10.1007/978-981-19-5209-8_22 ↗
15. A hybrid model integrating recurrent neural networks and the semi-supervised support vector machine for identification of early student dropout risk10.7717/peerj-cs.2572 ↗
16. Psathas, G., Chatzidaki, T.K., and Demetriadis, S.N. (2023). Predictive modeling of student dropout in MOOCs and self-regulated learning. Computers, 12.10.3390/computers12100194 ↗
17. Chi, Z., Zhang, S., and Shi, L. (2023). Analysis and prediction of MOOC learners’ dropout behavior. Appl. Sci., 13.10.3390/app13021068 ↗
18. Semi-supervised learning (chapelle, o. et al., eds.; 2006) [book reviews]10.1109/TNN.2009.2015974 ↗
19. A survey on self-supervised learning: Algorithms, applications, and future trends10.1109/TPAMI.2024.3415112 ↗
20. Kostopoulos, G., and Kotsiantis, S. (2021). Exploiting semi-supervised learning in the education field: A critical survey. Advances in Machine Learning/Deep Learning-Based Technologies: Selected Papers in Honour of Professor Nikolaos G. Bourbakis–Vol. 2, Springer.10.1007/978-3-030-76794-5_5 ↗
21. Self-labeled techniques for semi-supervised learning: Taxonomy, software and empirical study10.1007/s10115-013-0706-y ↗
22. Arazo, E., Ortego, D., Albert, P., O’Connor, N.E., and McGuinness, K. (2020, January 19–24). Pseudo-labeling and confirmation bias in deep semi-supervised learning. Proceedings of the 2020 International joint conference on neural networks (IJCNN), Glasgow, UK.10.1109/IJCNN48605.2020.9207304 ↗
23. Yarowsky, D. (1995, January 26–30). Unsupervised word sense disambiguation rivaling supervised methods. Proceedings of the 33rd Annual Meeting of the Association for Computational Linguistics, Cambridge, MA, USA.10.3115/981658.981684 ↗
24. Zou, Y., Yu, Z., Liu, X., Kumar, B., and Wang, J. (November, January 27). Confidence regularized self-training. Proceedings of the IEEE/CVF International Conference on Computer Vision, Seoul, Republic of Korea.
25. Li, M., and Zhou, Z.H. (2005, January 18–20). SETRED: Self-training with editing. Proceedings of the Pacific-Asia Conference on Knowledge Discovery and Data Mining, Hanoi, Vietnam.10.1007/11430919_71 ↗
26. Blum, A., and Mitchell, T. (1998, January 24–26). Combining labeled and unlabeled data with co-training. Proceedings of the Eleventh Annual Conference on Computational Learning Theory, Madison, WI, USA.10.1145/279943.279962 ↗
27. Semi-supervised learning by disagreement10.1007/s10115-009-0209-z ↗
28. Hady, M.F.A., and Schwenker, F. (2008, January 15–19). Co-training by committee: A new semi-supervised learning framework. Proceedings of the 2008 IEEE International Conference on Data Mining Workshops, Pisa, Italy.10.1109/ICDMW.2008.27 ↗
29. Zhou, Y., and Goldman, S. (2004, January 15–17). Democratic co-learning. Proceedings of the 16th IEEE International Conference on Tools with Artificial Intelligence, Boca Raton, FL, USA.
30. Wang, J., Luo, S.w., and Zeng, X.h. (2008, January 1–8). A random subspace method for co-training. Proceedings of the 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence), Hong Kong, China.
31. Co-training with relevant random subspaces10.1016/j.neucom.2010.01.018 ↗
32. Improve computer-aided diagnosis with machine learning techniques using undiagnosed samples10.1109/TSMCA.2007.904745 ↗
33. Random forests10.1023/A:1010933404324 ↗
34. Tri-training: Exploiting unlabeled data using three classifiers10.1109/TKDE.2005.186 ↗
35. Deng, C., and Guo, M.Z. (2006, January 13–17). Tri-training and data editing based semi-supervised clustering algorithm. Proceedings of the Mexican International Conference on Artificial Intelligence, Apizaco, Mexico.10.1007/11925231_61 ↗
36. A competency-based specialization course for smart city professionals10.58459/rptel.2024.19013 ↗
37. Panagiotakopoulos, T., Kotsiantis, S., Kostopoulos, G., Iatrellis, O., and Kameas, A. (2021). Early dropout prediction in MOOCs through supervised learning and hyperparameter optimization. Electronics, 10.10.3390/electronics10141701 ↗
38. Kizilcec, R.F., and Halawa, S. (2015, January 14–18). Attrition and achievement gaps in online learning. Proceedings of the Second (2015) ACM Conference on Learning@ Scale, Vancouver, BC, Canada.10.1145/2724660.2724680 ↗
39. How do we model learning at scale? A systematic review of research on MOOCs10.3102/0034654317740335 ↗
40. Crossley, S., Dascalu, M., McNamara, D.S., Baker, R., and Trausan-Matu, S. (2017). Predicting Success in Massive Open Online Courses (MOOCs) Using Cohesion Network Analysis, International Society of the Learning Sciences.
41. Educational data mining and learning analytics: An updated survey10.1002/widm.1355 ↗
42. Scikit-learn: Machine learning in Python
43. SSLearn: A semi-supervised learning library for Python10.1016/j.softx.2024.102024 ↗
44. A systematic analysis of performance measures for classification tasks10.1016/j.ipm.2009.03.002 ↗
45. Learning from imbalanced data10.1109/TKDE.2008.239 ↗
46. Chicco, D., and Jurman, G. (2020). The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC Genom., 21.10.1186/s12864-019-6413-7 ↗
47. Kostopoulos, G., Fazakis, N., Kotsiantis, S., and Dimakopoulos, Y. (2025). Enhancing Semi-Supervised Learning in Educational Data Mining Through Synthetic Data Generation Using Tabular Variational Autoencoder. Algorithms, 18.10.3390/a18100663 ↗
48. Tarvainen, A., and Valpola, H. (2017, January 4–9). Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results. Proceedings of the 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA.
49. Fixmatch: Simplifying semi-supervised learning with consistency and confidence
50. Berthelot, D., Carlini, N., Goodfellow, I., Papernot, N., Oliver, A., and Raffel, C.A. (2019, January 8–14). Mixmatch: A holistic approach to semi-supervised learning. Proceedings of the 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, BC, Canada.
51. Graph-based semi-supervised learning: A comprehensive review10.1109/TNNLS.2022.3155478 ↗