The authors developed and tested a model to identify cognitive presence in a small for-credit course. They then applied the model to a larger scale MOOC and found "moderate level agreement". The interesting result of the study was that "the classifiers for cognitive presence developed for one discipline was not sufficiently generic to use in the others." It could be because "the different vocabularies and collocations used in each discipline may confuse the machine learning algorithm since most of the classification features that we used were linguistic features."
Today: 6 Total: 89 [Direct link] [Share]
