• Home
• Issues
• 1462

MOOC Dropout Factor Discovery Study Based on Learning Behavior

Select Volume / Issue:

• Volume 6,Issue 1,Jan.,2019

Year:

2019

Type of Publication:

Article

Keywords:

Computer Assisted Learning, Communication Technology, Medical Education, Information Technology and Systems Analysis

Authors:

• Luo, Yong
• Xiao, Zhihao
• Li, Jianping
• Xie, Zheng

Journal:

IJIRES

Volume:

6

Number:

1

Pages:

130-138

Month:

January

ISSN:

2349-5219

BibTex:

@article{1462, author = "Yong Luo and Zhihao Xiao and Jianping Li and Zheng Xie", abstract = "MOOC is currently facing the challenge of high dropout rates. This article will mine the MOOC learning behavior data and explore the rules of online learning. Through studying the differences in online learning behaviors of various learners, we divided the MOOC learners into authenticator, practitioners, observer, visitor. Naive Bayes classification algorithm was introduced, and the classification of learners provides the basis for personalized learning. The Bayesian formula is used to mine the main factors of MOOC dropout from learning behavior data. Statistics show that watching course videos and doing exercises are the most important factors that causes students to give up. So, video recording and exercises design are still the most important works to improve the course retention rate. On the other hand, since most of the learners being not in order to obtain MOOC certificates, we need to design modules for them. Students can customize their own learning modules, which enhance the effectiveness of MOOC", issn = "23495219", journal = "IJIRES", keywords = "Computer Assisted Learning, Communication Technology, Medical Education, Information Technology and Systems Analysis", month = "January", number = "1", pages = "130-138", title = "{MOOC} {D}ropout {F}actor {D}iscovery {S}tudy {B}ased on {L}earning {B}ehavior", volume = "6", year = "2019", }

Abstract:

MOOC is currently facing the challenge of high dropout rates. This article will mine the MOOC learning behavior data and explore the rules of online learning. Through studying the differences in online learning behaviors of various learners, we divided the MOOC learners into authenticator, practitioners, observer, visitor. Naive Bayes classification algorithm was introduced, and the classification of learners provides the basis for personalized learning. The Bayesian formula is used to mine the main factors of MOOC dropout from learning behavior data. Statistics show that watching course videos and doing exercises are the most important factors that causes students to give up. So, video recording and exercises design are still the most important works to improve the course retention rate. On the other hand, since most of the learners being not in order to obtain MOOC certificates, we need to design modules for them. Students can customize their own learning modules, which enhance the effectiveness of MOOC

Full text: IJIRES_1462_FINAL.pdf Online version [Bibtex]