Semi-automatic method for grading a million homework assignments

Ben Lorica, , Oct 10, 2013
Commentary by Stephen Downes
files/images/stanford_codeweb_1_3.png

This is an interesting approach to the question of machine grading: "Clustering submissions along key metrics is a natural way to reduce the amount of work required. The hope is that homework submissions within the same cluster are similar enough, that feedback for one member can be propagated to the rest of the cluster... While this approach has yet to be used for grading, some instructors have used it to highlight common mistakes: clusters can also be used to identify common errors found in homework submissions."

Views: 0 today, 83 total (since January 1, 2017).[Direct Link]