I am working at a private university in Germany that is planning to use moodle as its primary e-learning plattform starting this year.
Consequently we do not have had any courses yet and no analytics models could have been trained using our data so far. As I understand those models correctly they apply logistic regression using only internal data from our institution. If this is the case there is no outcome yet observed (e.g. students that have completed any courses) such that analytics models cannot be applied in the first course ever held on a site, right?
Maybe there will be options for unsupervised models in the future?