ABSTRACT
In this article an interactive tool for supporting the generation of Predictive Models for edX MOOCs is presented. This tool is a modular and scalable application that, based on the data collected from a MOOC course, allows the simple application of a complete Data Science process, where Machine Learning algorithms are applied to generate predictive models which could be used to predict which learners have passed the course and therefore obtained a certificate. The presented tool is called "edX-MAS" and it helps the user to analyze and compare the prediction quality of different Machine Learning algorithms, allowing the possibility of exporting the results in order to do a more exhaustive analysis.
- C. Delgado Kloos, C. Alario-Hoyos, C. Fernández-Panadero, I. Estévez-Ayres, P. Muñoz-Merino, R. Cobos, J. Moreno, E. Tovar, R. Cabedo, N. Piedra, J. Chicaiza, and J. López. Proyecto eMadrid: MOOCs y Analítica del Aprendizaje. SIIE16, CEDI2016. Salamanca, Castilla y León, Spain.Google Scholar
- R. Cobos, A. Wilde, and Zaluska. Predicting attrition from Massive Open Online Courses in FutureLearn and edX. Comparing attrition prediction in FutureLearn and edX MOOCs. Proceedings of the LAK FutureLearn Worshop in the Learning Analytics and Knowledge 2017 Conference (LAK17), Canada.Google Scholar
- R. Cobos, S. Gil, A. Lareo, and F.A. Vargas. Open-DLAs: An Open Dashboard for Learning Analytics. LS 2016 Third (2016) ACM Conference on Learning Scale Edinburgh, Scotland Uk, pages 265--268. (Apr. 2016) Google ScholarDigital Library
- M.Leon, R. Cobos, K. Dickens, Su. White, and H. Davis. Visualising the MOOC experience: a dynamic MOOC dashboard built through institutional collaboration. eMOOCs 2016. 4th European MOOCs Stakeholders Summit. Pages 461--470, (Feb. 2016).Google Scholar
- L.A. Ruipérez Valiente, R. Cobos, P.J. Muñoz-Merino, A.l Andújar, and C. Delgado Kloos. Early Prediction and Variable Importance of Certificate Accomplishment. European MOOC Stakeholder Summit 2017 (eMOOCs 2017). Leganés, Madrid, Spain. (May 2017)Google Scholar
- V. Macías. Herramienta para el modelado predictivo en entornos educativos en línea.Universidad Autónoma de Madrid. Spain. Bachelor dissertation. (June 2017)Google Scholar
- I.D. Claros, R. Cobos, G. Sandoval and M. Villanueva. Creating MOOCs by UAMx: experiences and expectations. European MOOC Stakeholder Summit 2015. Mons, BelgiumGoogle Scholar
- M Leon, R. Cobos, and K. Dickens. Internal Perspectives of MOOCs in Universities. European MOOC Stakeholder Summit 2017 (eMOOCs 2017). Leganés, Madrid, Spain.Google Scholar
- J. Friedman, T. Hastie and R. Tibshirani. Additive logistic regression: a statistical view of boosting. Annals of Statistics 28(2), 2000. 337--407Google Scholar
- Y. Freund and R. E. Schapire. 1997. A decision-theoretic generalization of on-line learning and an application to boosting. Journal of Computer and System Sciences, 55(1):119--139. Google ScholarDigital Library
- J. Friedman. 2001. Greedy Function Approximation: A Gradient Boosting Machine. The Annals of Statistics, Vol. 29, No. 5 (Oct., 2001), pp. 1189--1232.Google ScholarCross Ref
- T. Chen, T. He and M. Benesty. 2016. Package 'xgboost': Extreme Gradient Boosting. Documentation available in https://cran.r-project.org/web/ packages/xgboost/xgboost.pdfGoogle Scholar
- C. Cortes, V. Vapnik. (1995). "Support-vector networks". Machine Learning. 20 (3): 273--297 Google ScholarDigital Library
- J.M. Keller, M.R. Gray and J.A. Givens. 1985. A fuzzy k-Nearest Neighbour algorithm. IEEE Transactions on Systems, Man and Cybernetics 15(4) 580--585Google ScholarCross Ref
Index Terms
- edX-MAS: Model Analyzer System
Recommendations
Towards the development of a learning analytics extension in open edX
TEEM '14: Proceedings of the Second International Conference on Technological Ecosystems for Enhancing MulticulturalityThe emergence of platforms to support MOOCs (Massive Open Online Courses) strengthens the need of a powerful learning analytics support since teachers cannot be aware of so many students. However, the learning analytics support in MOOC platforms is in ...
Predict and Intervene: Addressing the Dropout Problem in a MOOC-based Program
L@S '19: Proceedings of the Sixth (2019) ACM Conference on Learning @ ScaleMassive Open Online Courses (MOOCs) are an efficient way of delivering knowledge to thousands of learners. However, even among learners who show a clear intention to complete a MOOC, the dropout rate is substantial. This is particularly relevant in the ...
Tools to Support Self-Regulated Learning in Online Environments: Literature Review
Lifelong Technology-Enhanced LearningAbstractSelf-regulated learning (SRL) skills are especially important in Massive Open Online Courses (MOOCs), where teacher guidance is scarce, and learners must engage in their learning process trying to succeed and achieve their learning goals. However,...
Comments