Answers are grounded in this paper and cite their sources.
We propose an automatic marking scheme for grading essay type assessment. The scheme is built on top of the BERT model. First, it obtains the meaning of the sentences in the students' answers and the specimen answer. Then, it uses a classifier to determine the similarity between the meaning of the students' answers and the specimen answer. Experiments show that the scheme achieves high accuracy when marking assignments of specific topics.
No takes yet. Share an insight, caveat, or question.
Ye et al. (2020) studied this question.