Again we are adhering to the guidelines from the Framework Model (section Methods). The methods suggested are concerned with amount and composition of test material. In these evaluations we have not been so much concerned with amount as with composition, but in reality a fairly large set of error examples are called for. We have concentrated on the composition, and followed the guidelines repeated here:.
Measuring precision and recall is done by running a grammar checker on the test material and counting the number of rejected and accepted items, respectively.