A Corpus Study on the Difference between MOOCs and Real Classes

Abstract

In this paper, we take a look at how the language of Online classes (MOOCs) differs from those of real classes. Three corpora were created for this analysis; MOOC corpus, Lecture Capture corpus, and Philosophy Lecture Capture. Three factors were used in the study: Formality, Sentiment analysis, vocabulary analysis. Formality score was used to understand how formal the text is. Sentiment categorization of words was used to realize the positivity of the words used in the classes and finally, top words used in corpora were analyzed to understand the usage. It was realized that the formality measure of real classes is slightly lower than online classes and professors use more positive words in real classes than online classes and the vocabulary usage is heavily under the influence of subject.