Learning analytics on coursera event data: a process mining approach
| Autor: | Mukala, P., Buijs, J., Leemans, M., van der Aalst, W., Caravolo, P., Rinderle-Ma, S. |
|---|---|
| Přispěvatelé: | Process Science |
| Jazyk: | angličtina |
| Rok vydání: | 2015 |
| Předmět: | |
| Zdroj: | Proceedings of the 5th International Symposium on Data-driven Process Discovery and Analysis (SIMPDA 2015), Vienna, Austria, December 9-11, 2015, 18-32 STARTPAGE=18;ENDPAGE=32;TITLE=Proceedings of the 5th International Symposium on Data-driven Process Discovery and Analysis (SIMPDA 2015), Vienna, Austria, December 9-11, 2015 |
| Popis: | Massive Open Online Courses (MOOCs) provide means to offer learning material in a highly scalable and flexible manner. Learning Analytics (LA) tools can be used to understand a MOOC's effectiveness and suggest appropriate intervention measures. A key dimension of such analysis is through profiling and understanding students' learning behavior. In this paper, we make use of process mining techniques in order to trace and analyze students' learning habits based on MOOC data. The objective of this endeavor is to provide insights regarding students and their learning behavior as it relates to their performance. Our analysis shows that successful students always watch videos in the recommended sequence and mostly watch in batch. The opposite is true for unsuccessful students. Moreover, we identified a positive correlation between viewing behavior and final grades supported by Pearson's, Kendall's and Spearman's correlation coefficients. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|