Large scale analysis of open MOOC reviews to support learners’ course selection

作者：

Highlights：

• We conduct an analysis using the largest MOOC review dataset to date in literature.

• The review distribution is biased with most of them having 4 or 5 stars.

• We use topic modeling and sentiment analysis to support course selection.

• Results suggest a positive correlation between sentiment values and numeric ratings.

• We conduct a use case characterizing four different MOOCs using our models.

摘要

•We conduct an analysis using the largest MOOC review dataset to date in literature.•The review distribution is biased with most of them having 4 or 5 stars.•We use topic modeling and sentiment analysis to support course selection.•Results suggest a positive correlation between sentiment values and numeric ratings.•We conduct a use case characterizing four different MOOCs using our models.

论文关键词：Massive Open Online Courses,Natural language processing,Sentiment analysis,Recommendation systems,Online education

论文评审过程：Received 25 January 2022, Revised 29 July 2022, Accepted 3 August 2022, Available online 6 August 2022, Version of Record 13 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118400