This paper proposes a method for eval- uating grammatical error detection meth- ods to maximize the learning effect ob- tained by grammatical error detection. To achieve this, this paper sets out the following two hypotheses ?imperfect, rather than perfect, error detection max- imizes learning effect; and precision- oriented error detection is better than a recall-oriented one in terms of learning ef- fect. Experiments reveal that (i) precision- oriented error detection has a learning ef- fect comparable to that of feedback by a human tutor, although the u0002rst hypothesis is not supported; (ii) precision-oriented er- ror detection is better than recall-oriented in terms of learning effect; (iii)
展开▼
