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

The recent pandemic has changed the way we see education. During years, Massive Open Online Course (MOOC) providers, such as Coursera or edX, are reporting millions of new users signing up on their platforms. Though online review systems standard among many verticals, no standardized fully decentralized exist in MOOC ecosystem. In this vein, believe that there is an opportunity to leverage available open reviews order build simpler and more transparent reviewing systems, allowing really identify best courses out there. Specifically, our research analyze 2.4 million (which largest dataset used until now) from five different platforms determine following: (1) if numeric ratings provide discriminant information learners, (2) NLP-driven sentiment analysis textual could valuable (3) can topic finding techniques infer themes be important for (4) use these models effectively characterize MOOCs based reviews. Results show clearly biased (63% them 5-star ratings), modeling reveals some interesting topics related with course advertisements, real applicability, difficulty courses.