Examining Class Dropping with Data Mining

摘要

Dropping classes can have negative consequences on both students and on scheduling classes. If the class size decreases to less than the lower limit for running the class, that class will be cancelled. Administrators will have the burden of contacting students who are registered for that class to inform them that their class has been cancelled. In some cases, this action might have an impact on retention. The other unpleasant task is finding a class for the faculty whose class was cancelled. In this paper, class dropping will be analyzed using data mining. In particular, a classification model will be employed to study the profile of students who will most likely drop a class. Simulated data will be used for the analysis purposes.