ABSTRACT
With the booming of massive open online courses (MOOCs) in recent years, high drop-out rate and low completion rate remain major impediments. Analyzing the behavior of learners and exploring the reasons of dropout of course are of great significance for evaluating the teaching quality and improving the teaching effect. As educational data mining and machine learning rise, it is possible to explore the causes of high dropout rate and low completion rate on MOOCs. Here, we conducted analysis on data from five multi-disciplinary courses on icourse 163 platform and extracted more than 50 features for evaluating causes of withdrawing on MOOCs. A cause exploration method based on gradient boosting decision tree was implemented to extract the essential features that influence the learner's performance with the importance of each feature calculated. To get the best performed classification model, Cross-Validation was used in modeling and Grid Search was employed in parameter optimization. According to the importance of each feature, different subsets of features were selected and tested in turn. The results indicated nine features being the main factors affecting dropout. It could be expected to provide a guideline for a basis for evidence-based improvement of online education.
- A. Ramesh, D. Goldwasser, B. Huang, H. Daume III, and L. Getoor, "Learning latent engagement patterns of students in online courses," in Proceedings of the Twenty-Eighth AAAI Conference on Artificial Intelligence, 2014. Google ScholarDigital Library
- J. He, J. Bailey, B. I. Rubinstein, and R. Zhang, "Identifying at-risk students in massive open online courses," in Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence, 2015. Google ScholarDigital Library
- Girish K. Balakrishnan, Derrick Coetzee. Predicting student retention in massive open online courses using Hidden Markov Model. Technical Report Identifier: EECS-2013-109Google Scholar
- Jiajun Liang*, Chao LI, Li Zheng*. Machine Learning Application in MOOCs:Dropout Prediction{C}. The 11th International Conference on Computer Science & Education (ICCSE 2016) August 23--25, 2016.Google ScholarCross Ref
- Muñoz-Merino PJ, Ruipérez-Valiente JA, Alario-Hoyos C, et al. Precise Effectiveness Strategy for analyzing the effectiveness of students with educational resources and activities in MOOCs{J}. Computers in Human Behavior, 2015, 47(C):108--118. Google ScholarDigital Library
- Jiang Z, Zhang Y, Li X. Learning behavior analysis and prediction based on MOOC data{J}. Journal of Computer Research & Development, 2015(3):614--628.Google Scholar
- Hira S, Deshpande P S. Mining precise cause and effect rules in large time series data of socio-economic indicators{J}. Springerplus, 2016, 5(1):1625.Google ScholarCross Ref
- Ben Hoyle, Markus Michael Rau, Feature importance for machine learning redshifts applied to SDSS galaxies.Google Scholar
- Ma S, Li J, Liu L, et al. Mining combined causes in large data sets{J}. Knowledge-Based Systems, 2016, 92:104--111. Google ScholarDigital Library
- Muhammad R. Khan, Johua Manoj, Anikate Singh, et al. Behavioral Modeling for Churn Prediction: Early Indicators and Accurate Predictors of Custom Defection and Loyalty{J}. 2015, 92(3):677--680. Google ScholarDigital Library
- Li, W., Gao, M., et al. Dropout prediction in MOOCs using behavior features and multi-view semi-supervised learning. In 2016 International Joint Conference on Neural Networks (IJCNN), (pp. 3130--3137). IEEE.Google Scholar
- Chen T, Guestrin C. XGBoost: A Scalable Tree Boosting System{J}. 2016:785--794. Google ScholarDigital Library
- Rokach L. Ensemble-based classifiers{J}. Artificial Intelligence Review, 2010, 33(1--2):1--39. Google ScholarDigital Library
Recommendations
Supporting learners' self-regulated learning in Massive Open Online Courses
AbstractIn MOOCs, learners are typically presented with great autonomy over their learning process. Therefore, learners should engage in self-regulated learning (SRL) in order to successfully study in a MOOC. Learners however often struggle to self-...
Highlights- Learners struggle to regulate their learning in massive open online courses (MOOCs).
- A self-regulated learning (SRL) intervention was implemented in three MOOCs.
- Learners' SRL was measured with trace data variables.
- ...
Benefit and Cost Analysis of Massive Open Online Courses: Pedagogical Implications on Higher Education
There has been much research done on online learning including research on online educational activities and methods. The use of technology is gaining rising importance in higher education due to the benefits that it brings. In terms of adopting new ...
Benefit and Cost Analysis of Massive Open Online Courses: Pedagogical Implications on Higher Education
There has been much research done on online learning including research on online educational activities and methods. The use of technology is gaining rising importance in higher education due to the benefits that it brings. In terms of adopting new ...
Comments