Develop a ml data model using generated student grades to predict an expected grade. I think we should use the naive bayes model, since all the grades are independent of each other. We have to figure out how to create it with features that aren't just binary values, since each feature (course) has 14 values (A+,A,A-,B+,...,F). Also we shouldn't expect an exact match, if Jon took CS 1, CS 2, CS 3 and got a 'B+' in all of them, and Ryan took CS 1, CS 2 and got an 'A' in both of them, then an A should be predicted for Ryan taking CS 3. For now our base classifier (or a freshie/transfer) should just be the most probable grade given a professor and course.
Develop a ml data model using generated student grades to predict an expected grade. I think we should use the naive bayes model, since all the grades are independent of each other. We have to figure out how to create it with features that aren't just binary values, since each feature (course) has 14 values (A+,A,A-,B+,...,F). Also we shouldn't expect an exact match, if Jon took CS 1, CS 2, CS 3 and got a 'B+' in all of them, and Ryan took CS 1, CS 2 and got an 'A' in both of them, then an A should be predicted for Ryan taking CS 3. For now our base classifier (or a freshie/transfer) should just be the most probable grade given a professor and course.